VOLUME 8 ISSUE 17

JULY 2017

ISSN 1989-9947 Virtual Archaeology Review

Volume 8, Issue 17 (2017) Editorial Team

Editor-in-Chief José Luis Lerma, Universitat Politècnica de València, Spain

Associate Editors Editors Emeriti Juan Antonio Barceló, Universitat Autònoma de Alfredo Grande, Sociedad Española de Barcelona, Spain Arqueología Virtual (SEAV)-Virtual Archaeology International Network (INNOVA), Spain Fabio Remondino, Bruno Kessler Foundation (FBK), Italy Víctor Manuel López-Menchero Bendicho, Universidad de Castilla-La Mancha - SEAV, Spain Advisory Board Thomas E. Levy, Department of Anthropology & Qualcomm Institute, UC San Diego, United States Livio de Luca, French National Centre for Scientific Research (CNRS), France Sofia Pescarin, CNR ITABC Rome (IT), Italy Markus Reindel, Commission for the Archaeology of Non-European Cultures (KAAK), German Archaeological Institute (DAI), Germany

Editorial Board Areti Damala, University of Strathclyde, United Orla Murphy, University College Cork, Ireland Kingdom Eva Pietroni, National Research Council, Institute Inés Domingo Sanz, ICREA, Universitat de of Technologies Applied to Cultural Heritage Barcelona, Spain (CNR ITABC), Italy Michael Doneus, University of Vienna, Austria Daniel Pletinckx, Visual Dimension bvba, Belgium Stephen Fai, Carleton University, Canada Laia Pujol Tost, Universitat Pompeu Fabra, Spain Francisco R. Feito Higueruela, Universidad de Fulvio Rinaudo, Politecnico di Torino, Italy Jaén, Spain Mario Santana Quintero, Carleton University, Andreas Georgopoulos, National Technical Canada University of Athens, Greece Jin Shang, Tsinghua University, China Luis A. Hernández Ibáñez, Universidade da Minna Angelina Silver, University of Oulu, Finland Coruña, Spain Juan Carlos Torres, Universidad de Granada, David Hernández López, Universidad de Castilla Spain La Mancha, Spain Geert J. Verhoeven, LBI for Archaeological Arivaldo Leão de Amorim, Universidade Federal Prospection & Virtual Archaeology, Austria de Bahia, Brazil

Editing Committee Matilde Balaguer Puig, Universitat Politècnica de Mª Ángeles Hernández-Barahona Palma, SEAV València, Spain (Secretary), Spain Rand Eppich, Architect. ICOMOS - CIPA Heritage Ángel Marqués Mateu, Universitat Politècnica de Documentation València

Published by Universitat Politècnica de València. Official journal of Spanish Society of Virtual Archaeology.

Virtual Archaeology Review, 8(17), 2017 ii Virtual Archaeology Review

Volume 8, Issue 17 (2017)

Contents

Delivering and using 3D models on the web: are we ready? 1-9 Roberto Scopigno, Marco Callieri, Matteo Dellepiane, Federico Ponchio, Marco Potenziani

Historical imprints and virtual representation issues in Mid-Byzantine Acrocorinth 10-19 Castle Demetrios Athanasoulis, Xeni Simou, Theodora Zirogianni

Rebuilding Palmyra virtually: recreation of its former glory in digital space 20-30 Ahmet Denker

Recreating a medieval urban scene with virtual intelligent characters: steps to 31-41 create the complete scenario Ana Paula Cláudio, Maria Beatriz Carmo, Alexandre Antonio de Carvalho, Willian Xavier, Rui Filipe Antunes

Design of a virtual tour for the enhancement of Llíria’s architectural and urban 42-48 heritage and its surroundings José Miguel Maícas, María José Viñals

Digitization of religious artifacts with a structured light scanner 49-55 Alejandro Graciano, Lidia Ortega, Rafael J. Segura, Francisco R. Feito

Digital heritage training for historians in Europe: a local proposal 56-63 José-Antonio Ruiz Gil

Uso combinado de la fotografía digital nocturna y de la fotogrametría en los 64-74 procesos de documentación de petroglifos: el caso de Alcázar de San Juan (Ciudad Real, España) Víctor Manuel López-Menchero Bendicho, Ángel Marchante Ortega, Matthew L. Vincent, Ángel Javier Cárdenas Martín-Buitrago, Jorge Onrubia Pintado

Entorno 3D para el análisis y la recreación virtual de las actuaciones 75-83 arqueológicas en Cueva de la Cocina (Dos Aguas, Valencia, España) Agustín Díez Castillo, Alfredo Cortell Nicolau, Oreto García Puchol, Pilar Escribá Ruiz

Modelado 3D del cementerio de la misión salesiana Nuestra Señora de la 84-94 Candelaria (Río Grande, Tierra del Fuego, Argentina) Melisa A. Salerno, Daniel Hereñú, Romina C. Rigone, Ricardo A. Guichón

Virtual Archaeology Review, 8(17), 2017 iii

Virtual Archaeology Review, 8(17): 1-9, 2017 http://dx.doi.org/10.4995/var.2017.6405 © UPV, SEAV, 2015

Received: July 25, 2016 Accepted: March 15, 2017

DELIVERING AND USING 3D MODELS ON THE WEB: ARE WE READY? DISTRIBUCIÓN Y USO DE MODELOS 3D EN LA WEB: ¿ESTAMOS LISTOS? Roberto Scopigno*, Marco Callieri, Matteo Dellepiane, Federico Ponchio, Marco Potenziani Istituto di Scienza e Tecnologie dell’Informazione “Alessandro Faedo” (ISTI), CNR, via Moruzzi 1, 56124 Pisa, Italy. [email protected]; [email protected]; [email protected]; [email protected]; [email protected]

Abstract: Digital technologies are now mature for producing high quality digital replicas of Cultural Heritage (CH) assets. The research results produced in the last decade ignited an impressive evolution and consolidation of the technologies for acquiring high-quality digital three-dimensional (3D) models, encompassing both geometry and color. What remains still an open problem is how to deliver those data and related knowledge to our society. The web is nowadays the main channel for the dissemination of knowledge. Emerging commercial solutions for web-publishing of 3D data are consolidating and becoming a de-facto standard for many applications (e-commerce, industrial products, education, etc.). In this framework, CH is a very specific domain, requiring highly flexible solutions. Some recent experiences are presented, aimed at providing a support to the archival of archaeological 3D data, supporting web-based publishing of very high-resolution digitization results and finally enabling the documentation of complex restoration actions. All those examples have been recently implemented on the open-source 3D Heritage Online Presenter (3DHOP) platform, developed at CNR-ISTI. Key words: 3D digitization, virtual archaeology, cultural heritage, documentation, web-based 3D visualization, restoration

Resumen: Las tecnologías digitales están ahora maduras para producir réplicas digitales de alta calidad de valores activos del patrimonio cultural (CH). Los resultados de la investigación producidos en la última década han mostrado una evolución impresionante y una consolidación de las tecnologías para la captura de modelos digitales tridimensionales (3D) de alta calidad, que abarcan la geometría y el color. Lo que queda aún por resolver está relacionado con la forma de distribuir los datos y el conocimiento relacionado con la sociedad. La web es hoy en día el principal canal utilizado para divulgar el conocimiento. Las soluciones comerciales nuevas relacionadas con la publicación en la red de datos en 3D se están consolidando y convirtiendo en un estándar de facto para muchas aplicaciones (comercio electrónico, productos industriales, educación, etc.). En este escenario, el patrimonio cultural es un dominio muy específico, que requiere soluciones muy flexibles. Se presentan algunas experiencias recientes, destinadas a proporcionar un apoyo al archivo de los datos arqueológicos 3D, la publicación web de los resultados de digitalización de muy alta resolución que permiten finalmente la documentación de trabajos de restauración complejos. Todos estos ejemplos se han implementado recientemente en la plataforma 3D Heritage Online Presenter (3DHOP) de código abierto, desarrollada en el CNR-ISTI. Palabras clave: digitalización 3D, arqueología virtual, patrimonio cultural, documentación, visualización web 3D, restauración

1. Introduction panoramic/360º images; standard videos or omnidirectional videos) or the geometrical/shape Multiple digital technologies are experimented in Digital characteristics of a CH artefact or scene (using one of the Humanities (DH); those instruments are going to play an many 3D representation incarnations). important role in the study, dissemination and didactical A number of enabling technologies have been developed activities related to our Cultural Heritage (CH). These and matured in the last years to sample the artworks of technologies cover many different types of applications interest. Sampling devices are then paired by processing and are the results of intensive research in many scientific technologies, data optimization tools (e.g. providing domains (not just Computer Science but also Optics, simplification and multiresolution features to increase the Physics, Chemistry, etc.). More specifically, Computer usability of very high resolution samplings), efficient Science is offering many different opportunities to archival systems, search and retrieval facilities and, manage visual data at unprecedented levels of accuracy finally, tools for web presentation. and ease of use. The term visual data is intended here to encompass all digital visual representations we can adopt The latter domain is the main focus of this paper. to produce a useful sampling of either the visual Nowadays the web is the place consulted by everybody characteristics (by means of: standard 2D images; more searching for knowledge or digital content. This is sophisticated image media, such as RTI or definitely true for needs related to teaching or to

*Corresponding author: Roberto Scopigno, [email protected] 1

SCOPIGNO et al., 2017 personal interest (and in this case Wikipedia or systems technologies for producing high-quality digital 3D replicas providing collections of PowerPoint presentations are of CH artifacts. Most of these technologies are off-the- key resources). At the same time, consulting the web is shelf products, in many cases also at very low cost. They more and more common also to support academic study allow users to produce high-resolution models (with high- and research. Many CH applications would require the resolution we mean sampled models counting from 5 availability of visual data as well as tools to process million up to hundreds of million faces/points). Most of those data and to get insight from those data (Scopigno these technologies enable also very high-quality sampling et al., 2011). of the geometric data, with accuracies in the order of few tens of microns (at the small scale) or a few millimeters All the material produced by the many digitization projects (for technologies working at the large scale). in the last decade is extremely valuable raw digital content and has to be disseminated and shared with the CH applications require digitization should not focus only community of scholars and practitioners. This is to create digital models representing the shape, since we unfortunately still a wish rather than a reality. The amount also need a very good sampling of the color or, better, of of time and resources dedicated to 2D/3D digitization in the surface reflection properties of the object or the scene the last 10 years has been impressive, but only the tip of digitized. Several works have presented approaches for this iceberg has been exposed and offered to the sampling and mapping the color or the surface reflection community. The majority of the material produced properties (Lensch, Kautz, Goesele, Heidrich, & Seidel, remains hidden in the shelves or in the hard drives of the 2003; Callieri, Cignoni, Corsini, & Scopigno, 2008; people who did the digitization. There are a number of Dellepiane, Marroquim, Callieri, Cignoni, & Scopigno, reasons which explain the poor availability of good quality 2012) and those solutions are now part of many off-the- digital models: shelf digitization solutions.  Intellectual property rights (IPR) issues. Digitization Once we have a digital 3D sampling of our CH artwork, has still a non-negligible cost, usually payed by the the further issues are: how to encode efficiently those owner of the artwork; many CH institutions still want data; how to archive and made them accessible to the to keep control over the digital data, either to make community; and how to visualize them efficiently in the profit or to keep it safe themselves because they are framework of CH applications. worried by possible improper (commercial) use. The accuracy and density of the digital sampling is the  Several digitization efforts occur in the framework of intrinsic value of the reconstructed model, because that research projects, archaeological excavations or huge quantity of geometric data shows a high potential for restoration actions. Therefore, they are part of many applications. But, at the same time, very high- currently active researches/studies, usually still resolution data can be perceived as a problem by users unpublished and thus confidential. Consequently, (making the design of efficient computations kernels or open data access is often postponed after the end of visualizations more complex). Therefore, endorsing the specific project (and in many cases it becomes a methods able to produce a controlled granularity of the forgotten action at the end of the project). digital models is mandatory in CH applications.  Finally, even in the cases where the digitization Many years of research in computer graphics have been actions are leaded by very open and forward-minded instrumental in building an arsenal of technologies for curators/scientists and data dissemination is controlled surface simplification and for the construction perceived as an important goal, often the lack of of multiresolution encoding schemes and view- expertise or financial resources impede data dependent rendering modalities (Cignoni et al., 2004; disclosure. Cignoni et al., 2005; Borgeat, Godin, Blais, Massicotte, & Lahanier, 2005; Wimmer & Scheiblauer, 2006). Technological advances are bringing us good and ready Providing an interactive and fluid visualization of to use solutions which could solve easily at least the last complex models is a critical task for applications running barrier mentioned above. We have solid, performant and either on the web or on mobile platforms. Many view- easy to use solutions that allow producers to publish dependent rendering solutions have been presented, complex visual media content on the web and to share it. able to process a multi-resolution model and extracting The paper will first offer a brief review of some related frame-by-frame view-dependent representations that enabling technologies in Section 2 (solutions for CH fulfil the rendering quality and the performance digitization and data optimization/management). Later constrains. Some of these solutions, are now available some platforms that support the presentation on the web on the entire spectrum of platforms (from desktop of CH visual media are presented in Section 3, focusing computers to tablets and smartphones), demonstrating on the commercial platforms and the issues in web-based that we dispose of a common enabling technology able visualization. Section 4 presents the 3DHOP platform, to carry high-quality graphics to everyone and which enables flexible presentation of 3D models on the everywhere. An example is the Nexus library1, which web. Finally, Section 5 reports conclusions and a follows the approach proposed in Cignoni et al. (2005). discussion about future needs and forthcoming developments. Once we have the capability to generate good quality 3D models, the next group of enabling technologies are the 2. Enabling technologies for digitization ones needed to archive and access those data, hopefully and management of sampled CH data enriching the digital models with metadata and efficient search and retrieval functionalities. Some pioneering The progress with optical systems, digital photography work on this subject has been done in the EC and visual computing has produced a number of mature “3DCOFORM” project (http://www.3d-coform.eu/).

1 http://vcg.isti.cnr.it/nexus/

Virtual Archaeology Review, 8(17): 1-9, 2017 2

DELIVERING AND USING 3D MODELS ON THE WEB: ARE WE READY?

Integration of different media is a final issue. Digital 3D specification on how to render 3D data that web browsers models are just one of the media available and used to should implement. Hence, by incorporating the WebGL document the status and the beauty of CH assets. Images approach, modern web browsers are able to natively (both the standard ones and the more advanced 2D render 3D models by using the features of the 3D media such as RTI, panoramic or multispectral images) graphics hardware, without needing additional plug-ins or have an important role; video is a resource easier to extensions. Since WebGL is a low-level API, developing record and to distribute to users. The improved insight an application that uses WebGL is not an easy task and that can be gathered by the use of multiple media should requires considerable skills in graphics programming. be taken into account in the design of archives and Therefore, WebGL is not a solution by itself to the needs visualization systems. A future goal is to go beyond the and issues of the DH community. But, as we will present consolidated 3D graphics approach (just store and use in the following subsections, WebGL has been the the 3D media). This means designing and developing enabling basic layer for the development of several technologies able to link different media or to present interesting tools or resources. Figure 1 shows an example them in a coordinated or integrated manner (Snavely, of a web system developed using WebGL. Seitz, & Szeliski, 2006; Brivio et al., 2013; Messaoudi, Manuel, Gattet, De Luca, & Véron, 2014; Leoni et al., 3.1. Commercial platforms for publishing 3D 2015; Potenziani et al., 2015). content on the web The introduction of WebGL has ignited the development of many different approaches for managing 3D content on the web, both at the academic and commercial level. Some of these experiences have produced more sophisticated and higher-level libraries (Behr, Eschler, Jung, & Zöllner, 2009). In other cases, commercial services have been developed to support the easy publication and visualization of 3D content on the web.

3.1.1. Sketchfab An excellent representative of a commercial platform is Sketchfab2, considered today the de-facto standard for publishing 3D content on the web. Sketchfab provides a 3D model viewer based on WebGL working either on any mobile/desktop webpage or on VR headsets. It enables the user to move freely around or inside the 3D scene (a) (b) using the mouse or a touch-based manipulation. In Figure 1: An example of the visual presentation provided by the addition to static 3D models, the viewer is able to play and Cenobium system (Corsini et al., 2010): a single window control interactive 3D animations. Finally, users can enables the side-by-side visualization of a) a high-resolution enable the VR mode to make the model viewable in VR image and b) a 3D model, related to two capitols from the headsets. Cefalù cloister (Sicily, Italy). The 3D viewer is used to present the models uploaded on the Sketchfab website, but can also be embedded on 3. Presenting 3D models on the web external websites, notably on Facebook or on any The delivery of 3D content through the web started to be personal web site. supported with a considerable delay with respect to other Publishing content by means of Sketchfab is very easy. digital media such as text, still images, videos and sound. This tool follows the approach introduced by YouTube: Early approaches proposed for publishing and visualizing after registration and activation of an account, an upload 3D data on the web (e.g. VRLM, X3D) had a major webpage allows users to insert a few data on the specific disadvantage: they confined 3D data to a specific 3D data file and to upload it. All processing and visualization tool, implemented as a plugin (i.e., binary conversion is done in an automatic manner on the executable modules external to the hosting browser) that Sketchfab platform. Sketchfab users can choose to make had to be explicitly installed by users. This approach was their 3D model files available for download under Creative definitely improper for the CH community, where potential Commons license. users are usually not information and communication technology (ICT) experts and where the appearance of a Sketchfab is an excellent system, providing a very high- blank screen corresponding to a request of installation of performance and high quality rendering engine. The only a piece of software frequently discourages the user from limitations of this platform are the lack of flexibility (there further exploration of the data. is only one incarnation of the viewer, which follows a very clean design but follows the requirements of a very wide The appearance of the WebGL standard in 2009 set of possible applications). Therefore, it is not possible (Khronos Group, 2009) was a fundamental change. to configure or modify the Sketchfab viewer to support WebGL is the latest born component of the OpenGL specific CH needs. ecosystem, and it is modeled as a JavaScript Application Programming Interface (API) that exposes a one-to-one Moreover, Sketchfab supports only a mono-resolution mapping to the OpenGL ES 2.0 specifications for model (it enables the use of a few levels of detail, just to embedded systems. WebGL provides therefore a switch between interactive and not interactive mode, but

2 https://sketchfab.com/

Virtual Archaeology Review, 8(17): 1-9, 2017 3 SCOPIGNO et al., 2017 it does not support a real multiresolution approach). To 3.2. Managing data complexity on the web optimize download transmission times Sketchfab uses a lossy compression method when transferring the Existing commercial platforms for publishing 3D content geometry. Most of the CH models published on on the web in most cases are designed to present 3D Sketchfab are drastically simplified (geometry is usually objects by supporting a high-quality visual appearance at most around 1 million triangles) and most of the detail rather than keeping a precise geometric representation. is encoded in the associated texture map (in other In 3D graphics this means to replace as much possible of words, any user can enable the wireframe rendering the model geometry with textures. mode in the viewer settings to check how geometry is This choice is justified by aesthetical requirements: the usually low-res). quality of rendering is a very important aspect in public presentations/publishing, and the adoption of texture- based representation and rendering modes provides excellent visual results (when the current 3D data processing technologies are used in a correct manner). At the same time, the selection of a texture-based encoding is also justified by practical difficulties in handling complex 3D data (the data size of accurate 3D models implies excessive disk space requirements and transmission times, poor standardization and elevated intrinsic heterogeneity). These constraints become even more pressing for web environments (with limited computing resources and server disk space, and unknown available bandwidth). The design choice of giving priority to textured 3D models Figure 2: A snapshot of a 3D model published on the Sketchfab copes well with the needs of many application fields, repository (3D model with hotspots and additional information). where relatively geometrically-coarse 3D models can be used without problems. Conversely, CH applications Recently Sketchfab has revised its license policy to cope often need to rely on geometrically complex 3D models. with CH institutions or users, creating an entire section High-resolution digitized geometries are often essential to of its website for museum and cultural collections, and convey correct information to final users, for both adding “hotspots” functionalities to its basic 3D viewer. technical uses (documentation, catalogues, restoration, Figure 2 shows an example of a CH model with study and measurement), and dissemination purposes interactive annotations published online using (didactical uses, virtual museum). Therefore, here are a Sketchfab. number of cases where low-res textured models are not a proper approach, thus requiring platforms able to Sketchfab reached a major impact on the CH domain. The support the efficient transmission and visualization of very number of models available on the Sketchfab archive dense 3D models. depicting cultural items3 is now quite large and it increased with a very good speed in the last couple of An efficient management of large geometry-based 3D years. The quality of the material available is still uneven, models on the web can be achieved by porting the Nexus but it includes several good quality models coming from multiresolution approach (briefly introduced in Section 2) both academic/institutional providers and over the WebGL layer. The view-dependent rendering practitioners/amateurs. engine of the Nexus system has been ported from C++ to JavaScript, revealing that JavaScript performances are 3.1.2. Smithsonian Museum X3D not a major limitation due to the minimal processing required by the Nexus library (that requires at run time just The Smithsonian Museum X3D visualizer4 is another a simple traversal of the multiresolution hierarchy). The commercial solution for the visualization of 3D CH models limitations of the WebGL API with respect to the more (we tag it commercial since it has been designed by complex desktop OpenGL API are not critical here, due to AutoDesk to fulfil the requirements of the Smithsonian the very basic OpenGL features required by Nexus. Museum). It enables the visualization of 3D digitized However, some extensions and modifications have been specimens as well as their integration with other types of introduced while implementing the WebGL version of data. X3D shows a higher configurability level than Nexus, enhancing it with a more aggressive solution for Sketchfab, since it has been clearly designed to cope with the compression of the geometric 3D data and by the diversity of the CH artefacts stored in a museum. It is adopting a streaming approach over HTTP. The possible to attach additional information to the models, to combination of these two components enables an create a virtual "tour" on the objects moving among the efficient and fast data transfer even in the case of very various hotspots available on the virtual replica, and also high-resolution models. to provide tours featuring several objects, grouped following a common concept. While efficient and visually This multiresolution engine (Nexus on WebGL) is the pleasant, the viewer is currently a closed project, owned technology we are using in most of the CNR-ISTI by AutoDesk and in-development with the Smithsonian projects (e.g. the Cenobium system shown in Figure 1 museum, so its use is restricted and extendibility to other and the 3DHOP platform described in the following use cases/requirements is unclear. section).

3 https://sketchfab.com/categories/cultural-heritage 4 http://3d.si.edu/

Virtual Archaeology Review, 8(17): 1-9, 2017 4

DELIVERING AND USING 3D MODELS ON THE WEB: ARE WE READY?

4. The 3DHOP platform Pursuing the best compromise between rendering and streaming performances, the core engine of the viewer 3DHOP (Potenziani et al., 2015), acronym for 3D Heritage has been upgraded several times. Thanks to these Online Presenter, is a platform designed to cope with the improvements now 3DHOP can easily handle (in a light needs of the DH community. 3DHOP simplifies the multiresolution compressed format) both high-resolution creation of interactive visualization webpages and meshes and point clouds (tens of millions enables to display high-resolution 3D models with intuitive triangles/vertices and over). Concerning appearance user interaction/manipulation. Moreover, using 3DHOP management, it supports both textures-based and per- the 3D resources can be deeply connected with the rest vertex color encodings. The texture-based encoding of the webpage elements (see the example in Figure 3). works either on single-resolution models or on the usual multiresolution encoding; we provide an extended Nexus conversion tool able to produce a multiresolution texture-encoded model from an input high-resolution textured mesh. At the same time, several changes and extension have concerned the viewer usability and the interaction with the 3D scene: a revised user interface, support touch and multi-touch displays, a scene-related event handler and additional interaction trackballs, just to name a few. Furthermore, from the first version to date, also the 3DHOP toolset has been enhanced with several new tools: full screen feature, geometric hotspot tool, Figure 3: A screenshot of a web presentation built using selective model visibility, measurement suite, object 3DHOP (presenting a single 3D model with hotspots and transparency feature, and the cut-planes sectioning additional information). tool.

The most interesting characteristics of the 3DHOP framework are as follows:  It can work with extremely complex 3D geometries, using a streaming-friendly multiresolution scheme (the Nexus library, already introduced in Section 2 and 3);  It has been designed to work with web environments. So, thanks to the use of declarative-style scene creation and exposed JavaScript functions, it provides a satisfactory level of ease of use, focusing on developers with a basic background in web programming (but at same time remaining simple to approach for all other professionals); (a)  It is provided with a modular structure, composed by a number of basic building blocks for creating interactive visualizations, each one including a set of defaults variables, configurable, and equipped with a comprehensive documentation. 3DHOP is based on the WebGL subset of HTML5 and thus it works without the need of plugins on most modern browsers and on all platforms. The downloadable package, together to a detailed documentation, a series of tutorials (How-Tos) and a Gallery of examples, is available for consultation and test at the project official website5.

3DHOP has been released as open source software (GPL (b) licence) in April 2014. From then to date the framework has been constantly enriched with tools and features. It Figure 4: A couple of different layouts examples provided by 3DHOP. In a) is a layout for single object presentation, with the began its life as a simple 3D viewer able to handle only 3D scene equipped with the basic toolbar plus interactive triangular meshes with per-vertex color and equipped with hotspots linked to informative popup windows; in b) we show a a basic 4 buttons toolbar (home, zoom in, zoom out, light layout for the management of a collection of objects, with the control). Today 3DHOP is a much more complex and 3D viewer equipped with the basic toolbar, an informative panel flexible platform. on its right, and a slider selector over the collection of 3D models on its bottom.

5 http://3dhop.net

Virtual Archaeology Review, 8(17): 1-9, 2017 5 SCOPIGNO et al., 2017

However, it is important to stress that 3DHOP is not a browser, implemented using WebGL and appearing in a universal platform able to support any possible standard web page (see Fig. 5). application or visual communication project, but it is a framework designed to deal with specific needs. It is an Moreover, the new features allow for further ideal tool to visualize high-resolution single objects personalization of the page: it is now possible to change (especially with dense models coming from 3D scanning). the navigation paradigm and the style of the page. 3DHOP a very good choice for quickly creating interactive Moreover, new tools (i.e. for creating cut-through sections visualization for either a single or a collection of models and for taking point-to-point measurements) have been (as shown in Figure 4). made available, and they can be added to the visualization page. Conversely, 3DHOP is not suited to manage complex scenes made of several low-poly objects (this is a common case when working with CAD, procedural or hand-modelled geometries). Additionally, thanks to its exposed JavaScript functions, 3DHOP integrates extremely well with the rest of the webpage. The ideal situation is having the logic of the visualization scheme in the page scripts, and use 3DHOP for just the 3D visualization. Finally, 3DHOP has been designed with different levels of access, to be as straightforward as possible for the simpler cases but, at the same time, able to provide enough configurable features to support the huge variability of Cultural Heritage artworks and applications. (a) 4.1. Supporting automatic web publishing In the framework of the ARIADNE project (which concerns the development of infrastructures for the archaeology domain), the need for easy and automatic tools providing web publishing of visual media emerged as user request in a workshop held on 2013. The result of this workshop is the ARIADNE Visual Media Service (Ponchio, Potenziani, Dellepiane, Callieri, & Scopigno, 2016), a resource providing easy publication and presentation on the web of complex visual media assets. It is an automatic service that allows users to upload visual media files on an ARIADNE server and to transform them into an efficient web format, making them (b) ready for web-based visualization. The user is asked only to fill up a small form and to upload the raw file. All processing required to transform the data in a web- compliant and efficient format is done in an automatic manner by our server. The service supports three types of visual media: high- resolution images, Reflection Transformation Images (RTI), and high-resolution 3D models. The Visual Media Service6 was released on January 2015, has been extended in January 2016 and supports now the web publication and browsing of:

 High-resolution 2D images (input images are converted in a multi-resolution format and can be (c) browsed in real time, zooming in and out); Figure 5: Screenshots from the Ariadne Visual Media Service:  Reflection Transformation Images (RTI), also known a) The service landing web page; b) Visualization of an as Polynomial Texture Maps (PTM) images, i.e. uploaded RTI of a cuneiform tablet; and c) Visualization of a 3D dynamically re-lightable images (Mudge et al., 2008); model of a cinerary urn.  3D models (triangulated meshes, point clouds and 4.2. Supporting restoration with a web platform textured models). While all these tools are mostly devoted just to For each media type, we support automatic conversion to visualization, a recurring request from the community is to an efficient multiresolution representation, offering data use the web platform to access 3D data and work on it in compression, progressive transmission and view- a collaborative environment. Following this trend, CNR- dependent rendering. Each data type has a specific web- ISTI is currently contributing to an on-going project

6 http://visual.ariadne-infrastructure.eu/

Virtual Archaeology Review, 8(17): 1-9, 2017 6

DELIVERING AND USING 3D MODELS ON THE WEB: ARE WE READY? concerning the restoration of the fountain of Netpune7, a archival and retrieval. The 3D model is therefore a central complex monument in Bologna, Italy, which is an early component of the information system. work by Giambologna, completed about 1567. The digitization of the fountain was done by staff of the This fountain is undergoing a complex study to assess University of Bologna as one of the first actions of the its conservation conditions, that includes a large number project. CNR-ISTI, together with colleagues of the of scientific and visual analysis by a group of experts and University of Bologna, designed and implemented an restores leaded by ISCR (Istituto Superiore per la information system aimed at storing all the data gathered Conservazione ed il Restauro) and University of during the phases of analysis and restoration. This Bologna. system includes a database, designed specifically to The idea is that all the technical documentation for the support the specific data organization of this project, diagnostic phase, planning and execution of the which is interoperating with the 3D model of the fountain. restoration will be gathered in an online system, and Any user can query the information system or also browse everything will be geo-referenced on high-resolution 3D the 3D model of the fountain, using a simple web models (also available in streaming through the online interface. The 3D model becomes a sort of spatial index tool). The aim was not only to document the status of the to the information stored in the information system. For artwork before the restoration but also to support data example, all documents or photographs related to a

(a)

(b) (c) Figure 6: Some snapshots from the Neptune information system: a) The starting page, showing the entire monument and the access to the information system sub-components; b) After selection of a sub-element of the statue, the corresponding 3D model and all the related information can be browsed by the user; and c) Creation of a polygonal region over the head of the statue, which will become a single element of one of the conservation status reliefs.

7 http://nettuno.comune.bologna.it/

Virtual Archaeology Review, 8(17): 1-9, 2017 7 SCOPIGNO et al., 2017 specific point or component of the fountain can be stored The web can go beyond data archival and visualization; it in the system by linking them to this specific position over could also become the domain where programs were the skin of the 3D model; therefore, all data will be geo- running over data (we are making some experiments in referenced by means of links to the 3D model. Some this direction with the web-based version of MeshLabJS, images of the system (currently in testing phase) are http://www.meshlabjs.net/). presented in Figure 6. The DH domain is extremely rich and complex: CH The system is supporting the production of the reliefs professionals use many different types of visual media documenting the status of conservation. Instead of using in their study and analysis process, or for didactical an independent 2D drafting system (e.g. AutoCAD) or a purposes. In the near future, the consolidated approach Geographic Information System (GIS) system, the where we analyze visually each single dataset and Neptune information system allows the operators to media, will be replaced by a cross-evaluation draw the reliefs directly on the 3D model, creating either methodology resembling more the usual human point-based, polylines or 2D regions (see Fig. 6c) which approach on the real workbench. The single object are then stored in the relational database of the browser should be replaced by tools able to link and information system. cross-analyze different types of knowledge and data (each one linked to the virtual clone). The usual linear The system represents a quite complex artwork, 3D approach to present knowledge on a CH artwork, typical scanned at very high accuracy. 3D data management is in textual descriptions, shows all its limitations. We need implemented on top of 3DHOP technology and supports new systems able to break the linear approach, the web-based interactive manipulation and analysis of a supporting sophisticated capabilities for integrating and huge model (in total, around 600 million triangles). creating correlations among different media.

5. Conclusions Acknowledgements This paper has presented the status of the technology for The research leading to these results has received web-based publishing and visualization of visual media funding from the EU 7th Framework Programme (mostly focusing on CH 3D models). The web is nowadays (FP7/2007-2013) under grant agreement no. 654119 perceived as the main channel of publication and (EC "PARTHENOS" project) and EU H2020 dissemination of knowledge. Technology is now ready for Programme (“EMOTIVE: EMOTIve Virtual cultural sophisticated uses of the web (Messaoudi, Manuel, Gattet, Experiences through personalized storytelling”, De Luca, & Véron, 2014). The web is thus becoming the H2020-SC6-CULT-COOP-08-2016) under grant main dissemination and sharing platform for 3D content, as agreement no. 727188. we shown with several examples in this paper.

References Behr, J., Eschler, P., Jung, Y., & Zöllner, M. (2009). X3DOM: A DOM-based HTML5/X3D integration model. Proceedings of the 14th International Conference on 3D Web Technology (Web3D '09) (pp. 127–135). Darmstadt, Germany. http://doi.org/10.1145/1559764.1559784 Borgeat, L., Godin, G., Blais, F., Massicotte, P., & Lahanier, C. (2005). GoLD: Interactive display of huge colored and textured models. ACM Transactions on Graphics, 24(3), 869–877. http://doi.org/10.1145/1073204.1073276 Brivio, P., Benedetti, L., Tarini, M., Ponchio, F., Cignoni, P., & Scopigno, R. (2013). PhotoCloud: Interactive remote exploration of joint 2D and 3D datasets. IEEE Computer Graphics and Applications, 33(2), 86–96. http://doi.org/10.1109/MCG.2012.92 Callieri, M., Cignoni, P., Corsini, M., & Scopigno, R. (2008). Masked photo blending: Mapping dense photographic data set on high-resolution sampled 3D models. Computers & Graphics, 32(4), 464–473. http://doi.org/10.1016/j.cag.2008.05.004 Cignoni, P., Ganovelli, F., Gobbetti, E., Marton, F., Ponchio, F., & Scopigno, R. (2004). Adaptive tetrapuzzles: Efficient out-of-core construction and visualization of gigantic multiresolution polygonal models. ACM Transactions on Graphics, 23(3), 796–803. http://doi.org/10.1145/1015706.1015802 Cignoni, P., Ganovelli, F., Gobbetti, E., Marton, F., Ponchio, F., & Scopigno, R. (2005). Batched multi triangulation. Proceedings of the IEEE Symposium on Visualization (Info Vis '05) (pp. 20–214). Minneapolis, Minnesota, USA. http://doi.org/10.1109/VISUAL.2005.1532797 Corsini, M., Dellepiane, M., Dercks, U., Ponchio, F., Keultjes, D., Marinello , A., …, Wolf, G. (2010). CENOBIUM - Putting together the romanesque cloister capitals of the mediterranean region. B.A.R.-British Archaeological Reports International Series, 2118, 189–194. Retrieved from http://vcg.isti.cnr.it/Publications/2010/CDDPCKMSSW10/ Dellepiane, M., Marroquim , R., Callieri, M., Cignoni, P., & Scopigno, R. (2012). Flow-based local optimization for image- to-geometry projection. IEEE Transactions on Visualization and Computer Graphics, 18(3), 463–474. http://doi.org/10.1109/TVCG.2011.75

Virtual Archaeology Review, 8(17): 1-9, 2017 8

DELIVERING AND USING 3D MODELS ON THE WEB: ARE WE READY?

Khronos Group. (2009). WebGL - OpenGL ES 2.0 for the Web. Retrieved from https://www.khronos.org/webgl/

Lensch, H., Kautz, J., Goesele, M., Heidrich, W., & Seidel, H.-P. (2003). Image-based reconstruction of spatial appearance and geometric detail. ACM Transactions on Graphics, 22(2), 234–257. http://doi.org/10.1145/636886.636891 Leoni, C., Callieri, M., Dellepiane, M., Rosselli Del Turco, R., O'Donnell, D., & Scopigno, R. (2015). The dream and the cross: Bringing 3D content in a digital edition. Journal on Computing and Cultural Heritage (JOCCH) - Special Issue on Best Papers from Digital Heritage 2013 , 8(1), 5:1–5:21. http://doi.org/10.1145/2686873 Messaoudi, T., Manuel, A., Gattet, E., De Luca, L., & Véron, P. (2014). Laying the foundations for an information system dedicated to heritage building degradation monitoring based on the 2D/3D semantic annotation of photographs. Proceedings of the Eurographics Workshop on Graphics and Cultural Heritage. Darmstadt, Germany. http://doi.org/10.2312/gch.20141315 Mudge, M., Malzbender, T., Chalmers, A., Scopigno, R., Davis, J., Wang, O., …, Barbosa, J. (2008). Image-based empirical information acquisition, scientific reliability, and long-term digital preservation for the natural sciences and cultural heritage. Eurographics 2008 - Tutorials. Crete, Greece. http://doi.org/10.2312/egt.20081050.T2

Ponchio, F., Potenziani, M., Dellepiane, M., Callieri, M., & Scopigno, R. (2016). ARIADNE visual media service: easy web publishing of advanced visual media. Proceedings of the 43rd Annual International Conference on Computer Applications and Quantitative Methods in Archaeology (CAA 2015) (pp. 433–442). Siena, Italy. Retrieved from http://archaeopress.com/ArchaeopressShop/Public/ download.asp?id={77DEDD4E-DE8F-43A4-B115-ABE0BB038DA7} Potenziani, M., Callieri, M., Dellepiane, M., Corsini, M., Ponchio, F., & Scopigno, R. (2015). 3DHOP: 3D heritage online presenter. Computers & Graphics, 52, 129–141. http://doi.org/10.1016/j.cag.2015.07.001 Scopigno, R., Callieri, M., Cignoni, P., Corsini, M., Dellepiane, M., Ponchio, F., & Ranzuglia, G. (2011). 3D models for cultural heritage: Beyond plain visualization. IEEE Computer, 44(7), 48–55. http://doi.org/10.1109/MC.2011.196 Snavely, N., Seitz, S., & Szeliski, R. (2006). Photo tourism: Exploring photo collections in 3D. ACM Transactions on Graphics, 25(3), 835–846. http://doi.org/10.1145/1141911.1141964 Wimmer, M., & Scheiblauer, C. (2006). Instant points: Fast rendering of unprocessed point clouds. Proceedings of the 3rd Eurographics / IEEE VGTC Conference on Point-Based Graphics (pp. 129–137). Boston, Massachusetts, USA. http://doi.org/10.2312/SPBG/SPBG06/129-136

Virtual Archaeology Review, 8(17): 1-9, 2017 9

Virtual Archaeology Review, 8(17): 10-19, 2017 http://dx.doi.org/10.4995/var.2017.5959 © UPV, SEAV, 2015

Received: June 15, 2016 Accepted: April 13, 2017

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTH CASTLE HUELLAS HISTÓRICAS Y REPRESENTACIÓN VIRTUAL EN EL CASTILLO BIZANTINO MEDIO DE ACROCORINTO Demetrios Athanasoulisa, Xeni Simoub,*, Theodora Zirogiannic

a Hellenic Ministry of Culture/ Cyclades Ephorate of Antiquities, Epameinonda 10, 10555 Athens, Greece. [email protected]

b University of Patras, Department of Architecture,University Campus, 26500 Rio Patras, Greece. [email protected]

c Independent architect engineer, 20400 Xylokastro, Greece. [email protected]

Abstract: Fortified architecture of medieval times in Greece is a research field with several challenges for the scholars such as the absent of plans and written sources. The current study investigates how digital documentation methods can contribute to the extraction of archaeological information from the actual material sources -the various building traces. This relation between survey and interpretation will be presented through the case of Acrocorinth Castle. The survey of a smaller area of the fortification between the 2nd and 3rd defence line of Acrocorinth was executed in order to create a three dimensional (3D) representation for the Middle-Byzantine Phase of the castle. The project was implicated in the frame of the creation of the web-platform ecastles.culture.gr for the promotion of fortified architecture in Peloponnese under the Hellenic Ministry of Culture (25th Ephorate of Byzantine Antiquities, Argolid Ephorate of Antiquities). In the present study the surveying method and the step by step graphic reconstruction will be initially presented. Subsequently, it will be analysed how the digital survey led to the gradual modulation of the castle's historic form and how the scenarios of building phases' evolution were made. Key words: virtual archaeology, virtual reconstruction, data acquisition, documentation of cultural heritage, 3D reconstruction

Resumen: La investigación de la arquitectura medieval en Grecia es un área que presenta varios retos para los investigadores, ya que no existen mapas ni información escrita. El estudio actual investiga cómo los métodos de documentación digital pueden contribuir a la extracción de la información arqueológica a partir de las fuentes materiales existentes –las varias trazas de construcción. Esta relación entre documentación e interpretación se presenta mediante el caso del Castillo de Acrocorinto. La documentación de una pequeña área de la fortificación entre la 2a y 3a línea de defensa de Acrocorinto se ejecutó para crear una representación tridimensional (3D) de la fase bizantina media del castillo. El projecto se desarrolló en el marco de la creación de la plataforma web ecastles.culture.gr con el objeto de promocionar la arquitectura fortificada en el Peloponeso bajo el Ministerio de Cultura Helénico (25th Ephorate of Byzantine Antiquities, Argolid Ephorate of Antiquities). En el presente studio, se presenta el método de documentación y los pasos para la reconstrucción grafica. Adicionalmente, se analizará cómo la documentación digital condujo a la modulación gradual de la forma historica del castillo y cómo las diferentes fases de construcción fueron evolucionando. Palabras clave: arqueología virtual, reconstrucción virtual, adquisición de datos, documentación del patrimonio cultural, reconstrucción 3D 1. Introduction tools. The survey of Acrocorinth, part of the broader digitisation planning (Athanasoulis et al., 2015), was It is rather common place that despite the development of executed in summer 2015 by the architect-conservator advanced techniques for 3D representation, every survey writers Xeni Simou and Theodora Zirogianni during their is unique and demands the tailoring of suitable occupation in the aforementioned project (Athanasoulis, representation methodology. The survey approach is also Simou, & Zirogianni, 2016). Its main objective was to bonded with the specific aim of the survey. When working produce plans of the current condition of the monument in situ the architect-surveyor has to cope with complicated and to develop a 3D representation for a certain area problems of form, function, historic phases and (Fig. 1) of the fortified complex. The questioning that construction in order to produce not only accurate but also arised through the documentation process was the cause comprehensive measuring results. The questioning that is for the conducting of this essay, that aims to display the molding simultaneously with the survey is the key- interdependency of survey method and monument's element to unlock the monuments' tale and can be interpretation. confirmed with the numerical results of the surveying

*Corresponding author: Xeni Simou, [email protected] 10

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTH CASTLE

Figure 1: General view of Acrocorinth with indication of the survey area between the two Byzantine Defence Lines.

2. The case study Post-Byzantine times. The 3D representation concerns the Middle-Byzantine Acrocorinth's visualisation of the Acrocorinth Castle crowns the abrupt cliff that rises by area between the 1st and 2nd Byzantine defense lines (Fig. Ancient Corinth, a city of major importance in Ancient 2). The chronological period of the representation was Greek and Byzantine times. The selection of building the selected in order to restore castle’s peak-point city by the Isthmus was crucial for the flourishing of construction time. It was also chosen due to the relatively Corinth and played a critical role for the controlling of the big existing structural elements that permit the re- military passage ways of Peloponnese Peninsula and the imagination of the Middle-Byzantine phase. maritime routs on either side of the canal (Carpenter, Bon, & Parsons, 1936). During the ancient times from 7th century BC to the Roman destruction, Acrocorinth functioned as the Acropolis of the city (Carpenter, Bon, & Parsons, 1936). At the extended fortification project of Justinian in Peloponnese (6th century AD), Acrocorinth was largely developed as a part of the overall fortification plan of the Peloponnese. In the Middle Byzantine Period (843-1210) intensive urban and fortification development is taking place, following in a large scale the ancient wall understructure (Athanasoulis, 2014). During the post Byzantine times, Acrocorinth's conquerors apply their own alterations to the fortification according always to war technology trends, alterations that form in a big scale the contemporary view of the castle (Athanasoulis, 2014). After the Frankish Principality acquisition, Venetians and Figure 2: Area between 1st and 2nd Byzantine Defence Lines. Ottomans execute major fortification works (Andrews, 2006). The castle was finally abandoned after the Greek Independence War and its last siege in 1823. 3. Selection of survey method The castle in its contemporary form consists of three The documentation methodology can be described as a successive fortification lines with three gates respectively complete geometric documentation comprising the and runs a perimeter of 3000 m. The current formation of triptych historical–architectural–bibliographic documenta- the castle, as previously mentioned, is a long-time tion (Ioannides, Georgopoulos, & Scherer, 2005), process, a huge Byzantine fortification based on an resulting in a full archive of 2D current condition plans that ancient basis that has been extensively transformed in were later on elaborated to produce 3D model. The

Virtual Archaeology Review, 8(17): 10-19, 2017 11

ATHANASOULIS et al., 2016 survey was developed with a combination of techniques, based essentially on total station rather than using extensively (a single more advanced) 3D laser scanning technology. This choice was driven by the specificities of the monument itself and the goal of the survey that demanded a very analytic interpretation of the complex's edifice. Beyond the educational goal of 3D virtual representation, the survey efforts of Acrocorinth focused on creating analytic plans of the existing situation, as part of an integrated project for the enhancement and restoration of the huge complex of Acrocorinth. The acquisition of measurement data was, in essence, determined by the nature, the complexity and the location of the monument. Acrocorinth consists of a complicated multilayered construction with considerable altitude difference and hardly accessible areas or even places where the setup of a machine is impossible. Thus, tailoring of the survey method premised the solution of brain teasers connected with the peculiarities of the castle's structure and location. The digital survey was (c) based on multiple techniques (Fig. 3) and required developing a methodology of seamless combination, filling gaps and avoiding overlapping (Remondino et al., 2009). The EDM (Electronic Distance Measurement) technology with a Leica Total station TS06 model was extensively used for the measurement of general plans after sketching of the preliminary recording plans (ground plans, sections, unfolded elevations of the fortification walls) where a detailed list of surveyed points was kept. The blind areas such as openings and details were completed by detailed hand-measured drawings (Fig. 3).

(d) Figure 3: Different survey data: a) hand measurements; b) total station 3D points acquisition; c) rectification of images; d) laser scanning orthoimage (Archive of 25th ΕΒΑ, previous survey).

Then the image rectification processing for both inner and (a) outer walls followed, based on the survey points and sketches. Laser scanning data (Fig. 3) for geometrically interesting selected areas and interior spaces (El-Hakim et al., 2007) from previous and new surveys was exploited and linked to general plans by means of common control points. This laser scanning had been executed in order to capture both a tower's interior and a revealed building unit during excavation works in the last years. The laser scanning surveys were undertaken by colleagues of the 25th Ephorate of Byzantine Antiquities (EBA).

4. 3D reconstruction methodology Because castles pose complications in their presentation in digital virtual reality formats (Finan, 2012) there was a necessity of tailoring a personalised 3D methodology for (b) Acrocorinth.

Virtual Archaeology Review, 8(17): 10-19, 2017 12

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTH CASTLE

The intermediate step between the survey of the current previous building phases, if these were apparent or condition and the virtual representation of the Mid- subjacent to posterior constructions. Where the parapets Byzantine phase was the careful analysis of the were missing, the supplementation of the superstructure possessed material, either those that came from the was completed by keeping the general proportions of meticulous survey or those derived from indexing of similar battlements of the castle. historic archives, plans, representations and pictures. Doubtless, the demanding process of building phases Despite the fact that advanced 3D laser scanning identification as well as the material recognition maps technologies have been developed, the 3D model (Fig. 4) was essential in order to comprehend the construction followed a synthetic process. The adopted structural layers. That resulted to the discrimination of the 3D process was not such an automated technique as 3D initial Byzantine parts that remained intact through the data acquisition from laser scanning. During the 3D ages. modelling procedure new questions arised, related with areas that were neither visible nor accessible during the This investigation became more assiduous for peculiar 2D surveying and therefore not automatically resolved. elements or construction specificities that arised questions The modeller had to go back and give an answer to of original form and function. The deeper observation on questions of 3D spatial gaps. The 3D representation was structural state problems for instance could reveal the made in Autodesk 3ds Max 2015 environment after generator cause behind a possible collapse that happened inserting the 2D reconstruction plans (Fig. 5a). The due to instability or human factor and was responsible for system was using small number of polygons in order to extensive reinforcements on the masonry. be lighter and handy for touring options. For the rendering there was used a modified scaled pattern while for the Then followed the synthetic phase of creating 2D graphic understructure of ancient masonry mapping, the surfaces reconstruction plans. The representation was equally were unwrapped to create seamless corners. The final 3D based on the existing indicators on the masonry of the touring outcome (Fig. 5b) was made in Unity WebGL in castle and the assumptions created by comparisons with collaboration with a web developing company. Hot-spots similar elements.In order to reconstruct the parapets, the of archaeological information were introduced to the 3D. graphic restoration was completely following the traces of

Figure 4: Analysis process from survey plans to graphic reconstruction: a) survey plans of the existing condition; b) rectification of façades; c) building phases' diagram; d) graphic reconstruction plans.

Virtual Archaeology Review, 8(17): 10-19, 2017 13

ATHANASOULIS et al., 2016

resulted in a generally accurate 3D model with maximum local errors of 8-10 cm.

(a)

Figure 6: South elbow tower traces of lateral walls.

6. Interpretation of survey traces The interpretation of the monument can be perceived as the intellectual process of turning source material into conclusions (Pletinckx, 2008). Here follows, a descriptive apposition of specific details of the survey process of the castle. The interpretation methodology that has been developed based on specific building traces will be demonstrated in order to show how the aforementioned (b) combinational digital survey efforts contributed to the overall reconstruction attempt. Figure 5: Modelling process: a) in Autodesk 3ds Max 2015 environment; b) final 3D tour in Unity with pop-up info windows operated by Dataverse Ltd. 6.1. Continuity and extension assumptions Although relics standing scattered are incomprehensive, 5. Accuracy issues whether we investigate them as a part of a whole they Accuracy in surveys dealing with visualisation purposes become valuable. The continuity of a construction can be is depended on different specialists' approach on heritage confirmed by the exact measurements and can lead to the (architects, archaeologists, computer vision experts and drawing of the overall form. This was for instance the case engineers) and the accuracy requirements should be in the gate of the southeast corner that was an obvious defined based on the survey purpose and targeted public remnant of a previously existing elbow tower (Fig. 6). The (Santana Quintero, Neuckermans, Van Balen, & Jansen, accurate measurement and sketching of the small traces 1999). Additionally, scale and accuracy are important of the lateral walls made it possible to outline the exact properties of geometric documentation that should be form of the tower (Fig. 7). By prolonging these wall traces carefully defined at the outset, before any action on the to the right direction a rectangular-shape tower resulted. monument (Georgopoulos & Ioannidis, 2004). 6.2. Completion with analogies Due to the fact that Acrocorinth's survey drawings were meant to be used for further conservation purposes the Extracting inferences through analogy of an already acceptable error for the general plans was defined from known sample is actually a tool and a whole philosophical 1 cm up to 2.5 cm. These accuracy requirements approach, fundamental for archaeological science. determined to a certain degree the survey methodology. Missing elements can be supplemented based on Laser scanning precision and 3D data acquisition was analogy that is a form of reasoning whereby the identity necessary in excavated areas of tower interiors, while at of unknown things or relations is inferred from those that the same time photogrammetric 2D projected views are known (Ashmore & Sharer, 2010). were sufficient for the curtain walls survey and In the previous case of the elbow tower, the overall shape representation. could not be assumed by field traces. Completely missing However, taking into account the educational purpose of elements such as the exact lenght of lateral walls, the the model and because a major part of the masonry was overall height and the crenelations shape were filled in by graphically reconstructed based on analogies, the comparison with the standing towers that had been well accuracy requirements of the 3D modelling were lower. preserved. The general width was proportionally In order to reduce modelling polygons for the web estimated to the existing towers. Parapets' form and engine 3D virtual reconstruction, some negligible height was completed by following an acceptable height, inclinations and deformations were simplified that defined by the neighboring curtain wall traces.

Virtual Archaeology Review, 8(17): 10-19, 2017 14

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTH CASTLE

correlated with a building phase or element. In the case of the 2nd Byzantine Gate that is a solid mass of great complexity due to the continuous historical transformation, the detailed measuring was decisive for the crystallisation of a convincing assumption. There was a vertical joint in the back, only visible in the elevation. The precise measurement of this joint showed that it was corresponding to the lateral walls of the adjacent tower. An observational visitor could possibly recognise different building phases. But how these were correlated and what happened actually? The as-built plans were made, and then followed observation of the parallels, features matching (Fig. 9a) and simultaneous comparison of other prototypes. This led to the deduction that the Gate was initially a two-story tower flanked on either side by two towers in contact (Fig. Figure 7: South elbow tower: a) ground plan of existing 9b). The gate was protected by a portcullis mechanism. condition; b) ground plan of graphic reconstruction.

6.3. Peculiar structural elements recognition Another possible method for the interpretation is the recognition of peculiar indications on the construction that could possibly confirm a construction hypothesis. The southern tower (Fig. 8) was not fully excavated due to the existence of posterior adjacent historical structures on its back. The entrance to the southern tower was blocked by posterior additions. The significance of these structures did not permit to continue the archaeological excavations and the entrance investigation. Therefore there have been some unresolved questions on its initial form. Survey revealed that a suspicious corner extruding from the parapets (Fig. 8a) could solve the accessibility problem to the superstructure of the tower. The metrical reconstruction was matching with the hypothesis of a straight line staircase attached to the towers platform. The precise measurement led to the design of a staircase that fits ideally to the given height. (Fortunately, excavation works in the tower north to the gate showed the existence of a similar staircase.)

(a)

Figure 8: South Tower's survey plans with special indication of the staircase corner trace: a) ground plan; b) transversal section.

6.4. Corresponding joints or parallels (b) Survey data can be decisive for extracting valuable Figure 9: 2nd Byzantine Gate Tower: a) ground plan and historic information. The survey’s accuracy often defines elevations indicating joints and parallels; b) final 3D whether joints and parallels are corresponding and representation.

Virtual Archaeology Review, 8(17): 10-19, 2017 15

ATHANASOULIS et al., 2016

In posterior period, when artillery platforms for cannons were made, the building was extended, crenellations were changed, the function of the floor was repealed and the built staircase leading to the floor was replaced by a ramp.

6.5. Rejecting and confirming scenarios When other methods are inadequate the conclusions can come by a "reductio ad absurdum" technique. The interpretation needs the selection of the best solution between the range of suitable analogies (Ascher, 1961). The recognition of the most preferable solution, however, demands the development of several reconstruction hypothesis and the careful rejection of them one by one. This method was used in the re-thinking of the 1st Gate in Byzantine times (Fig. 10) that was the most difficult and complicated riddle. Although Byzantine masonry conceives a big part of the gate's structure, the tower was always an object to successive modifications. Figure 11:1st Byzantine Gate 2D survey: a) Byzantine parts in elevations, internal view; b) ground plan with indications of fragmented Byzantine surface.

Figure 10: 1st Byzantine Gate external view, 25th EBA Archive.

The recognition of byzantine traces (Fig. 11) was not sufficient for understanding the buildings' evolution but it was the first step. Survey revealed strange wall Figure 12: 1st Byzantine Gate 3D graphic reconstruction. inclinations, extrusions, discontinuities, hidden spaces, and incomprehensive areas, overlapping of architectural elements, badly executed and misleading previous 7. General reconstruction view restorations. Some of the questions arised were: What The detailed surveying, interpretation and the 3D was the initial shape? What was the function? Where was representation of the area between the west defence lines the entrance to the floor? gives a quite concrete image of Acrocorinth during the The process of graphical representation of different Mid-Byzantine period. The two defence lines, exploiting initial versions tried to give answers to these questions the existing ancient masonry traces were articulated in a and led directly to the creation of multiple possible way of creating an amphitheatrically organised platform scenarios that were finally one by one rejected for protected by towers and tall curtain walls (Fig. 13). different reasons each, mainly based on construction Ancient and early Byzantine material had been reused in (ex. not corresponding ground plan elements to critical places of the construction as spolia. façades, misleading deformations, typologically The towers of the 2nd Byzantine defence line, accessible impossible assumptions, chronologically impossible through the inner yard had a typical two vaulted units' solutions, etc.). interior configuration. The defender could enter to the The most reasonable scenario prevailed in the end and level of loop holes through a built staircase descending to resulted to the representation of a two-storied the interior of the tower from its back. The towers' roof passageway-tower with a two-unit first floor accessible platform and the curtain walls' parapets were accessible from the west platform (Fig. 12). Later on parts of the through built staircases attached in parallel to the façades were restored and other reinforced with extra backside of the walls. layers while a new entrance opened and disrupted the The gates were opened as walkways in two-storied tower's shape. towers (Gate towers) and were protected by strong

Virtual Archaeology Review, 8(17): 10-19, 2017 16

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTH CASTLE

Figure 13: Overall 3D view of the area between 1st and 2nd fortification lines in the Middle-Byzantine period. adjacent provisions. The 1st Gate was positioned Without the photogrammetric methods the creation of slantingly to the 2nd. This layout of the fortification allowed precise thematic maps such as building phases, or the maximum active defence of the entrance to the castle material discrimination would be a laborious and time by the defenders (Athanasoulis, 2014). consuming procedure. The level of detail of the picture with a high resolution gives second-level valuable The alterations made in the adventurous posterior information. In order to comprehend the edifice, periods were mainly executed in order to reinforce the photogrammetry worked complementary to the eye- construction and to install new war-technology inspection that sometimes was extremely difficult due to provisions such as cannon platforms. Comparing to the physical obstacles and restrictions. northern part that was extensively reconstructed in the Post-Byzantine period, in the southern part the ancient While elaborating the survey data in the office, and Mid-Byzantine material was better preserved. The fragmented site elements such as disrupted masonry elements subjected to successive interventions were could be reconnected due to the digital drawings' definitely the gate-towers. precision and could confirm the reconstruction hypothesis. On the contrary, cases like eye-illusions 8. Assessment of survey's contribution to misleading to false drawings were rejected. In addition to the interpretation and reconstruction this, the survey in unexcavated areas helped to identify surface elements, thus developing a surface investigation process for further archaeological excavations. We should admit the fact that digital methods provide like “Virtual archaeology complements perfectly no others an indisputable accuracy, crucial in deducing documentation and conservation efforts and even can act conclusions about building's form. It is also evident that as an integration activity to bring all information together selection of survey methodology is a difficult and in a structured way” (Pletinckx, 2009). In general all the determinant task for interpretation. As shown in the case acquired data from this survey can be stored and study of Acrocorinth, this surveying methodology and connected with older and new surveys for future use when accuracy was indeed valuable in analysing the necessary. construction and indicating its borders. It was important for interpreting whether a wall declination for instance or For the 3D reconstruction the team employed a synthetic a special structural element has been the result of initial and subtractive approach for the Middle-Byzantine construction, of later restoration, of weathering factors or impression of Acrocorinth Castle. It was actually a has been caused by human destruction. rebuilding of selected historic elements, essential to make

Virtual Archaeology Review, 8(17): 10-19, 2017 17

ATHANASOULIS et al., 2016 an integrated view of the monument. Unlike material continue the study, correcting errors and suggesting new reconstruction that is not an endless decision-making archaeological interpretations (Forte, 2011). process, certainly dependent on space limitations, inadequacy of archaeological evidence, and restoration Acknowledgements priorities (stability of the monument, safety reasons, new function, preservation for safeguarding the initial material, This project was a part of a wider collective attempt for full reconstruction for educational reasons, symbolic the digital enhancement project of castles of Arcadia, restoration, etc.), the 3D archaeological reconstruction Argolid and Corinthia co-funded by NSRF and Greek has to be as complete as possible in order to cover virtual Ministry of Culture. Thus, we owe special thanks to the tour options. In such a modelling of a specific historic personnel of Argolid and Corinthia Antiquities and period, the scholars had to give answers to the totality of especially Ephorate of Argolid Antiquities Dr the monument represented. As a result the assumptions archaeologist Alcestis Papademetriou and civil engineer were inevitable. When there is no evidence, 3D modelling Nikolaos Sidiropoulos. is based on hypothesis on different degrees of likelihood The acknowledgments should also particularly refer to the (Georgopoulos, 2014). On the other hand the 3D archaeologists Antonios Georgiou, Anna Sfika and reconstruction was a non-destructive way with great Vasiliki Klotsa for their contribution to the understanding potentials. The collected and elaborated data can be of the castle. Finally we should attribute gratitude to reused for the proposal of different alternatives in the architect-engineer Gregorios Kousoulos and civil- areas where the historic scenarios are weak. Current 3D engineer Athanasios Xourafis as well as all the surveyors representation is a tessel in scientific mosaic knowledge. that participated in previous enhancement projects in Then, the possibility of transmission of a cybernetic model Acrocorinth and employees that worked for the should allow the future communities of scientists to excavation and restoration projects.

References Andrews, K. (2006). Castles of the Morea. USA: Princeton. Ascher, R. (1961). Analogy in Archaeological Interpretation. Southwestern Journal of Anthropology, 17(4), 317–325. http://doi.org/10.1086/soutjanth.17.4.3628943 Ashmore, W., & Sharer, R. (2010). Discovering Our Past: A Brief Introduction to Archaeology. USA: McGraw-Hill Education. Athanasoulis. (2014). The Castle of Acrocorinth and its Enhancement Project. Ancient Corinth: Hellenic Ministry of Culture. Athanasoulis, D., Georgiou, A., Simou, X., Sfika, A., Klotsa, V., Zirogianni, T., …, Deligianni, E. (2015). Bridging monuments through digital repository and graphic reconstruction methodologies - The digital enhancement project of Argolid, Arcadia and Corinthia castles, Greece. Proceedings International Conference Digital Heritage. Granada: Spain. http://doi.org/10.1109/DigitalHeritage.2015.7413846 Athanasoulis, D., Simou, X., & Zirogianni, T. (2016). Historical traces' interpretation and virtual reconstruction-The case of Acrocorinth. In J. Lerma, & M. Cabrelles, Proceedings of the 8th International Congress on Archaeology, Computer Graphics, Cultural Heritage and Innovation “ARQUEOLÓGICA 2.0” in Valencia (Spain), Sept. 5-7, 2016 (pp. 288–290). Valencia: Editorial Universitat Politècnica de València. http://dx.doi.org/10.4995/EMERGE2014.2014 Carpenter, R., Bon, A., & Parsons, W. A. (1936). Corinth: Volume 3. The Defenses of Acrocorinth and the Lower Town. USA: Harvard University Press. El-Hakim, S., Gonzo, L., Voltolini, F., Girardi, S., Rizzi, A., Remondino, F., & Whiting, E. (2007). Detailed 3D modelling of castles. International Journal of Architectural Computing, 5(2), 199–220. http://doi.org/10.1260/147807707781515011

Finan, T. (2012). 3D castle reconstruction as interpretive modelling: The medieval Lough Cé Project. Château Gaillard 2 : L’origine du château médiéval, actes du colloque de Rindern, (pp. 45–49). Rindern, Allemagne.

Forte, M. (2011). Cyber archaeology: Notes on the simulation of the past. Virtual Archaeology Review, 2(4), 7–18. http://dx.doi.org/10.4995/var.2011.4543 Georgopoulos, A. (2014). 3D virtual reconstruction of archaeological monuments. Mediterranean Archaeology and Archaeometry, 14, 155–164. Georgopoulos, A., & Ioannidis, C. (2004). Photogrammetric and Surveying Methods for the Geometric Recording of Archaeological Monuments. Athens, Greece: FIG Working Week 2004. Ioannides, M., Georgopoulos, A., & Scherer, M. (2005). Standards in cultural heritage: the missing grammar for the digital documentation of the past. XX CIPA International Symposium. Torino, Italy. Retrieved from https://www.researchgate.net/publication/237252815_STANDARDS_IN_CULTURAL_HERITAGE_THE_MISSING_G RAMMAR_FOR_THE_DIGITAL_DOCUMENTATION_OF_THE_PAST

Virtual Archaeology Review, 8(17): 10-19, 2017 18

HISTORICAL IMPRINTS AND VIRTUAL REPRESENTATION ISSUES IN MID-BYZANTINE ACROCORINTHCASTLE

Pletinckx, D. (2008). Interpretation management: How to make sustainable visualisations of the past. Proceeding of the EPOCH Conference on Open Digital Cultural Heritage Systems. Rome, Italy. Retrieved February 25, 2016, from http://public-repository.epoch-net.org/rome/09%20Interpretation%20Management.pdf Pletinckx, D. (2009). Virtual archaeology as an integrated preservation method. Virtual Archaeology Review, 2(4), 33–37. http://doi.org/10.4995/var.2011.4545 Remondino, F., El-Hakim, S., Girardi, S., Rizzi, A., Benedetti, S., & Gonzo, L. (2009). 3D virtual reconstruction and visualization of complex architectures – The “3D-Arch” Project. 3D-ARCH. Trento: Italy. Retrieved February 5, 2016, from https://pdfs.semanticscholar.org/cf87/2fdcede5f3860ff7928e6d0aaa6e4c58e74d.pdf Santana Quintero, M., Neuckermans, H., Van Balen, K., & Jansen, M. (1999). Accuracy in affordable technology for three- dimensional documentation and representation of built heritage. Retrieved February 24, 2016, from https://lirias.kuleuven.be/bitstream/123456789/76035/1/vsmm99-santana-vanbalen-neuckermans-jansen.pdf

Virtual Archaeology Review, 8(17): 10-19, 2017 19

Virtual Archaeology Review, 8(17): 20-30, 2017 http://dx.doi.org/10.4995/var.2017.5963 © UPV, SEAV, 2015

Received: June 16, 2016 Accepted: April 20, 2017

REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE RECONSTRUCCIÓN VIRTUAL DE PALMIRA: RECREACION DE SU ANTIGUA GLORIA EN UN ESPACIO DIGITAL Ahmet Denker Faculty of Engineering and Natural Sciences, Istanbul Bilgi University, Eyup 34060, Istanbul, Turkey. [email protected]

Abstract: This paper addresses the role of the digital tools and methods in the preservation of cultural heritage. As the destruction of Palmyra made international headlines, digital tools emerged as a key tool to fight back against the total deletion of the heritage site from memories. Palmyra in Syria had been, with its Corinthian colonnades, theatre and splendidly built temples to ancient gods, source of inspiration and imagination for Western architecture. In this paper, the aim is reviving the lost reality of Palmyra by digitally reconstructing its “ghost images” from rubbles. The paper offers a glimpse of the grandeur and beauty of the ruins of Palmyra, none of which any longer remains in its entirety. 3D models of the most significant structures of Palmyra, including Temple of Bel and Temple of Baalshamin which have been levelled as a result of conflict based vandalism, as well as the Grand Colonnaded Street and the theatre are presented as “ghost images” through reconstructed 3D models. The focus has been in maintaining the accuracy and validity of the visualised data of the relics and environment of Palmyra, as they were once extant. Following the trauma caused by the destruction of Palmyra, these reconstructions revive in our memory the splendour of the city in the Greco-Roman Period. Key words: virtual archaeology, UNESCO world heritage site (WHS), 3D reconstruction, digital cultural heritage

Resumen: Este artículo trata el papel de las herramientas digitales y los métodos en la preservación del patrimonio cultural. Si bien la destrucción de Palmira propició titulares internacionales, las herramientas digitales han emergido como una herramienta clave para defenderse contra la supresión total de las memorias del sitio patrimonial. Palmira en Siria, ha sido con sus columnatas corintias, teatro y los templos de los antiguos dioses construidos espléndidamente, una fuente de inspiración e imaginación en la arquitectura occidental. Este artículo ofrece una vista rápida de la grandeza y la belleza de las ruinas de Palmira, ninguna de las cuales permanece hoy en día. Los modelos 3D de las estructuras más significativas de Palmira, como son el Templo de Bel y el Templo de Baalshamin, así como la gran columnata monumental y el teatro, que han sido severamente destruidos a causa de los conflictos vandálicos, se presentan como “imágenes fantasma” a través de los modelos reconstruidos en 3D. El enfoque ha sido mantener la precisión y la validez de la información visualizada de las reliquias y del ambiente de Palmira, tal y como eran antiguamente. Siguiendo el trauma causado por la destrucción de Palmira, estas reconstrucciones reaniman en nuestras memorias el esplendor de la ciudad en la periodo Grecorromano. Palabras clave: arqueología virtual, patrimonio mundial de la UNESCO, reconstrucción 3D, patrimonio cultural digital

1. Introduction Palmyra was known as “the Venice of the sands” and “the bride of the desert”. The cultural heritage which holds a significant part of the intellectual wealth of our information society has Palmyra used to contain well preserved temples which constantly been under threat of demolition. The threats were consecrated to ancient Mesopotamian gods, a of destruction have originated primarily, from local and colonnaded street with Corinthian columns, a theater, international dissents and conflicts. Throughout history and the valley of tombs. It had long been counted among war has always taken a toll on heritage as well as the most eloquent and stupendous ruined cities from the human lives. Damage by Talibans of the 1700 years old ancient world. As one of the outstanding cultural centers Buddhas of Bamiyan at a World Heritage Site (WHS) in of the world, it was also listed as a UNESCO’s WHS Afghanistan in 2001 was followed by the demolition of (Campbell, 2007). the antique city of Palmyra in Syria by ISIS (Islamic The city rose to prominence 2000 years ago during the State of Iraq and Syria) in 2005. reign of "the warrior queen", Zenobia, who resembled Pliny the Elder said: "Situated in a vast expanse of sand Cleopatra of Egypt in many ways. She had reigned like and renowned for its fertile soil and pleasant streams, Cleopatra and was defeated by Romans like her. Her the ancient city of Palmyra was a stopping point for regina concurred with an epoch when Romans began to caravans traversing the Syrian Desert" (Pliny the Elder, take interest in Palmyra, firstly for economical and lately Natural History 5.88.1). An oasis in the Syrian Desert, for political reasons. An abrupt rise of Roman interest for

*Corresponding author: Ahmet Denker, [email protected] 20 REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE oriental goods led to sudden growth in Palmyra to the sources are sought for filling the missing information. In size of one of the largest cities in the East. It linked the case of Palmyra, it is possible to reach a large Roman world to Mesopotamia; caravans were transiting volume of information. Textual and graphic descriptions Palmyra while crossing the Syrian Desert, and their governing the ancient architecture of Palmyra exist in transit brought wealth to the city. Its sudden and huge literature (Wiegand, 1932). Each demolished temple and prosperity was also put to an abrupt end in Roman the other landmarks of the city, colonnaded road, hands. From the day when Aurelian terminated theatre, etc. were to a great extend recorded. Zenobia’s rule of desert, the roads of commerce Recognition of the splendour of the ruins of Palmyra by diverted, and the fortune of Palmyra dwindled (Seyrig, th th 1950). travellers in the 17 and 18 centuries contributed greatly to the subsequent revival of classical As the latest act in the dark history of vandalism, ISIS architectural styles and urban design in the West. The destroyed Temple of Baalshamin (24 August 2015). It work of British explorer Robert Wood may be used as an was followed by Temple of Bel (31 August 2015), the instance (Wood, 1753). This work appeared subsequent Grand Colonnaded Street and the Arch of Triumph (4 to Wood’s visit to Palmyra in the course of a voyage he October 2015). They were unique monuments of a undertook with "two gentlemen whose curiosity had hybrid architectural style which blended Greco-Roman carried them more than once to the continent, canons with ancient Middle-Eastern architecture. Most of particularly to Italy." the landmarks of Palmyra are now gone. With their destruction and disappearance, irreplaceable treasures They thought that “a voyage, properly conducted, to the of the UNESCO WHS are lost to posterity. most remarkable places of antiquity, on the coast of the Mediterranean, might produce amusement and 3D computer graphics and virtual reality are the only improvement to themselves, as well as some advantage known means of remedy for our collective and to the public”, as he stated in the preface to his book. irreversible loss. Since 2015 there have been several The success of the book is primarily due to the drawings undertakings in order to revitalize the memory of the lost carried out by Giovanni Battista Borra. These were the splendor of Palmyra, by reconstructing the city in 3D. first accurate records of the monumental ruins of “Revive Palmyra“, “The New Palmyra“, “Ghost of Palmyra. Palmyra“, and “The Virtual Palmyra“ are some of titles given to the projects which seek to resurrect Palmyra. Wood tells how he joined them in the book as follows: "a “The Virtual Palmyra” was launched by the author and its fourth person in Italy, whose abilities as an architect and results were presented in a relatively confined way in the draftsman we were acquainted with, would be absolutely 8th ARCHEOLÓGICA 2.0 Conference (Denker 2016). necessary. We accordingly wrote to him (Giovanni Battista Borra), and fixed him for the voyage. The This paper forwards a more extended treatment on the drawings he made, have convinced all those who have subject and offers a glimpse of the grandeur and beauty seen them that we could not have employed anybody of the treasures of Palmyra, none of which any longer more fit for our purpose." remains in its entirety. 3D models of the most significant structures of Palmyra are put forward, including Temple With Giovanni Battista Borra’s drawings (Fig. 1) which of Bel, Temple of Baalshamin which have been levelled were engraved in 1753, Wood’s book has been the main as a result of conflict based vandalism, as well as the repository of information as regards to graphical data. Grand Colonnaded Street plus Great Tetrakionion and the Triumphal Arch. The 3D reconstruction of the theatre which had been the scene of numerous executions is also included.

2. Digital construction of virtual representations The project aims at reaching the following goals: 1. The 3D reconstruction of each of the lost edifices of Palmyra in as close form to their original as possible. 2. The placement of the individual models within the recontextualized 3D environment.

3. Piecing together the individual 3D models and the Figure 1: Drawing of the colonnaded street by Giovanni Battista 3D environment to establish a comprehensive virtual Borra. representation of the whole of Palmyra. Reaching these goals requires the availability and Another important source of graphical data has been the accessibility of the following data: drawings of Louis François Cassas (Cassas, 1900). Cassas had travelled to Palmyra in 1785. He made  Graphical and photographic data. several drawings of the ruins of Palmyra. His drawings  Archaeological survey data. are of indispensable value in providing the much needed graphical data about these ruins (Fig. 2).  Topographical data. Transporting this information into digital models may be When the data is missing and architectural elements are compared with Evans’ procedure of "translation from destroyed due to ravages of time and humankind, other drawings to buildings" (Evans, 1995). His proposition,

Virtual Archaeology Review, 8(17): 20-30, 2017 21 DENKER, 2017

which was the common sanctuary for all of the four tribes. It was almost a national cult centre and represented all of Palmyra (Schlumberger, 1971). Bel originated from the great god of Babylonia which shows before the Greco-Roman influence, inhabitants of Palmyra were culturally dependent on Babylon (Seyrig, 1950). Bel was identified as "Father God" of the universe like Zeus of the Greeks and Jupiter of the Romans. The cosmic nature of Palmyrean Gods is expressed in that it occurs in a deity trio in cult community with Yarhibol, an old God associated with the sun, and Aglibol, a lunar deity. Also Baalshamin, which as God of heaven, fertility, lightning and rain was equated with the Greco-Roman Dionysus/Bacchus. Bel was often portrayed together with Yarhibol and Aglibol Figure 2: Drawing of Temple of Baalshamin by Louis François representing the Sun and the Moon; the three gods Cassas. formed a triad. A 1st century AD relief in Louvre "there is only one communicant, and that is the drawing", Museum shows Bel between Aglibol and Yarhibol. inspires a digital process which ends up with the casting Unlike the Greco-Roman deities who were constantly in of the digital construction through a series of geometric quarrel, cooperation resided among the gods of projections of drawings. Palmyra (Gawlikowski, 1995). However, drawings of Giovanni Battista Borra and Louis The cult of the individual gods practiced organised priest François Cassas have some intrinsic limitations. Not processions. These cult processions were led by the everything can be arrived through them, e.g. texture, necessary liturgical actions as sacrifice and prayer and colour and light. Hence, efforts for digital construction of organised regular worship processions. The most virtual representations which start from the drawings of important sacrificial rites were the burning of incense in Giovanni Battista Borra and Louis François Cassas, small fire altars or the offering of fruits (Schmidt-Colinet, require absorption of supplementary data. 1995). Photos taken between 1867 and 1876 by Felix Bonfils, Palmyra’s temples were the remarkable examples of which provide the most complete visual record of monumental architecture which blended the Greco- Palmyra from the 19th century, have been another Roman architecture with Oriental architecture. The indispensable information source which enabled the hybrid elements of these temples demonstrated the realisation of this project. numerous cultures that frequently overlapped and inter- mixed in Palmyra. Temples of Palmyra were among the What engaged the greatest attention in this project has greatest architectural achievements of humankind until been what the monumental relics of Palmyra looked like they fell victim to vandalism. They drew on wealthy, in ancient times. We have tried to reconstruct their canonical forms stemming from both Greco-Roman and ancient state rather than their pre-demolition state. Ancient Near Eastern roots (Veyne, 2015). In an exhibition put up by the Smithsonian's Freer and 3. Temples of Palmyra Sackler galleries in 2015, the main theme was that the Unlike the other major cities of the eastern provinces of temples of Palmyra were a prime inspiration for the Neo- Rome where a single monumental temple to the patron classical architectural style in Britain and North America. god dominated the landscape (e.g. Temple of Artemis at They have had a direct influence on US architecture, Ephesus, Temple of Apollo at Didyma), there were including buildings such as the Capitol, the White House numerous temples in Palmyra. The reason for this lies and Monticello, the Virginia’s home of President Thomas behind the sophisticated religious structure of the city. Jefferson (O’ Brien, 2015). Although from their outward appearance they looked Greco-Roman the deities they housed were not. 4. Digital reconstruction Due to its semi-nomadic social structure which is There are several stages of digital reconstruction of an composed of individual branches rooting from the same ancient site. These include 2D drawing of antique stem there was no real Pantheon in Palmyra, that is to monuments in comparative dimensions; placing them in say no generally accepted gods hierarchy. According to relative positions and orientations on the site plan; the ancient written sources, there were four creating 3D graphical elements with surfaces composed commonwealth constituent tribes, each settled in a of a number of polygons. These surfaces are given different part of the city. Each tribe worshiped, besides photorealistic appearance through the use of appropriate its own local deities, gods of various origins. In addition texture maps and lighting. The placement of the light to indigenous Arab deities, they worshiped gods whose sources which generate the solar illumination and origins can be traced back to north or southern Syria, shadow effects in Syrian Desert were imperative for Arabia or Mesopotamia (Schmidt-Colinet, 1995). creating the right atmosphere. When rendered the reconstructed 3D virtual objects exhibit the Palmyra had four main temples which corresponded to characteristics of the real objects. these four commonwealth constituent tribes: the Temple of Baalshamin, for the "Lord of Heaven", the Temple of A variety of software tools have been implemented for Nabu, for the Mesopotamian god of wisdom and oracles the realisation of these phases in the Palmyra restitution like Apollo, the Temple of Allat for the goddess who can project. Autodesk AutoCAD® was employed for 2D be equated to Athena. Above all, was the big Bel Temple drawings of plans and elevations. The software used for

Virtual Archaeology Review, 8(17): 20-30, 2017 22 REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE the 3D reconstruction of Palmyra was Autodesk 3ds a heap of rubble (Reuters, 2016). The only part which Max®. In this project, when the virtually created 3D remained from the temple was the western doorway into objects were integrated according to the city plan, the file the sanctuary building (Fig. 4). was comprised of more than 1 million polygons and its size exceeded 2.5 GB. The environment surrounding the integrated 3D objects is important for creating the "realistic looking" scenes. It includes topography, vegetation, background images and sky.

5. Digital reconstructions of temples Out of the four major temples of Palmyra, Temple of Nabu and Temple of Allat were left alone by ISIS probably due to their low state of preservation compared to Temple of Bel and Temple of Baalshamin. The heart was ripped out of everyone with the news of destruction of the other two. The scale of devastation was immense Figure 4: The only part which remained from the Temple of Bel and the losses were irretrievable. is the western doorway into the sanctuary building (Photo by Maher Al-Mounes/AFP/DGAM) (Al-Mounes, 2016). Virtual reconstruction/recreation is the only available means for soothing the pain of the loss of these temples. Before blown up by ISIS in 2015, on the last day of We have aimed at the production of photo-realistic August, Temple of Bel was the most impressive and reconstructions and visualisations of Temple of Bel and renowned building in Syria. It was the vestige with which Temple of Baalshamin and present the results below. Palmyra was associated. It was consecrated to Bel, the chief of the Palmyrene Gods, the equivalent of Greek Zeus and Roman Jupiter. Its relic was preserved remarkably well and constituted the most outstanding part of the ruins. It used to exhibit the splendid synthesis of Greco-Roman and Ancient Near Eastern Architectures. The temple building was on a podium in the middle of an almost square open court (175 x 180 m). It was well framed by a portico of 30 m depth which meant that the outer dimensions of the temenos were extended to 205 x 210 m. There were Corinthian columns on the outside wall, and porticos had double rows of Corinthian columns. One exception to this was the west side where there was only one row. The temple stood in the court asymmetrically, facing the north. Its main axis tilted only by 5º from the North-South axis. The main axis of the temple divided the open court in two portions with a ratio of 3:2. Western and eastern Figure 3: 3D reconstruction phases and final "realistic looking” portions were approximately 105 m and 70 m deep, view of Temple of Bel. respectively (Figs. 5 and 6).

Lumion® was implemented for construction of the environment. As an example, the 3D reconstruction phases of Temple of Bel and its final view are shown in Figure 3.

5.1. Great Temple of Bel In the south-eastern part of the old city which was enclosed by the ancient wall, used to arise the main temple of Palmyra, Temple of Bel. It was one of the four main temples of Roman Syria, together with Temple of Jupiter in Damascus, Baalbek Temple, and the Jerusalem Temple built by Herod the Great. What made Bel unique was that it was the only one of those four which was preserved in shape and form.

Satellite pictures taken after Syrian government forces Figure 5: Shaded bird’s eye drawing of the Temple of Bel and retook the city, showed the entire structure collapsed in its temenos [in m].

Virtual Archaeology Review, 8(17): 20-30, 2017 23 DENKER, 2017

that access to the cella was not through a frontal door. The door was placed asymmetrically on the side wall which required the visitor to turn 90º in order to view the cult area (Fig. 9).

Figure 6: Shadowed flank elevation drawing of the Temple of Bel [in m].

At the level of stylobate the main temple building stands over a rectangular area of 29.52 x 54.6 m. Figure 9: Reconstructed flank-elevation of Temple of Bel (west- The cella was a rectangular building which was of 15.8 x side). 39 m. It was surrounded by one row of columns (Fig. 7). From the ancient inscriptions as well as the different architectural style of the colonnades surrounding the peribolos, it was understood that the temple building was older than its surrounding (Wiegand, 1932). In its outward appearance, the temple looked to derive from the canon of Hellenistic architecture (Fig. 10). The entrance to the inner court was from west side through a monumental propylaeum of 35 m width (Fig. 11). It occupied almost one-sixth of the front wall. Visitors were led to the gate through a majestic staircase. A protruding portico with 8 columns was decorating the front elevation. The isomorphic reconstruction of this temple was presented in (Seyrig et al., 1975). Figure 7: Plan of the main temple building [in m] (After Seyrig et al., 1975).

The order was Corinthian. The columnar arrangement was pseudo-dipteral. There were 8 columns at façades, 12 at the west flank and 14 at the east flank. Column height was 15.81 m, and base diameter was 1.33 m, the height of the building was 33.14 m (Fig. 8).

Figure 10: Reconstructed façade of Temple of Bel.

Figure 8: Flank Elevation of the main temple building [in m] (After Seyrig et al., 1975).

The asymmetry of columns at the flanks is a noteworthy aspect of Bel’s architecture. It was due to Figure 11: Reconstructed propylaeum of Temple of Bel.

Virtual Archaeology Review, 8(17): 20-30, 2017 24 REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE

The drawings of Propylaeum were given in Wood’s temple was destroyed by ISIS in August 2015. The 3D book. After Palmyra was recaptured, it was obvious that reconstructions of Temple of Baalshamin and its the damage to Bel -the most important monument of surroundings are shown in Figs. 16 and 17. Palmyra- was extensive.

5.2. Temple of Baalshamin Temple of Baalshamin was situated 200 m north of the Grand Colonnaded Street. It was facing east. The main temple building was situated in a semi-rectangular shaped temenos with a maximum length of 163 m and a maximum width of 58 m (Fig. 12).

Figure 14: Drawing of the flank elevation of Baalshamin [in m] (After Wiegand, 1932).

Figure 12: Shaded bird’s eye drawing of Temple of Baalshamin and its temenos [in m].

As with Temple of Bel, Temple of Baalshamin also exhibited hybridity in design. The Greco-Roman traits were demonstrated by its colonnaded precinct, prostyle, façade and tetrastyle structure (Collard, 1970). The main temple building was standing on a 10.6 m x 16.9 m stylobate (Fig. 13). The four free standing columns in the portico were finished in Corinthian order. They were 7.8 m high and standing on attic bases; the Figure 15: Shaded drawing of the portico of Baalshamin [in m]. total height of the building was 12.9 m (Fig. 14). Cella walls were decorated by Corinthian style flat plasters.

Figure 16: Reconstructed outer-view of Temple of Balshamin and its environment.

Figure 13: Plan of the main temple building [in m] (After Wiegand, 1932).

Alongside overall classical Greco-Roman appearance, it also embodied prominent Near Eastern motives. The most visible of them were the windows of cella (Figs. 14 and 15). Windows were also used in the Temple of Bell. These windows which did not exist in the Greco-Roman cannon signified that the Deity was inside. They also let more light inside the cella than usual. The Association for the Protection of Syrian Archaeology reported that the Figure 17: Reconstructed view of the temenos of Temple of Baalshamin.

Virtual Archaeology Review, 8(17): 20-30, 2017 25 DENKER, 2017

6. Digital reconstructions of other The beauty of the Arch of Triumph was so imposing that landmarks it became the symbol of Palmyra soon after the first illustrations of its ruins were published in the 18th century 6.1. Grand Colonnaded Street in Wood’s book (Wood, 1753). Grand Colonnaded Street, or decumanus, used to strike Tetrakionion of Palmyra which is believed to be the travellers by the distinctly Roman imperial aspect it constructed at the end of the 3rd century is one of the had bestowed upon city. It was another of Palmyra's most striking constructions of the city. It is a prominent great archaeological site with an architecture that fused echo of the columnar architecture of the Roman world. It Greco-Roman techniques with local traditions and is composed of four tetrapylons which appear overall as Persian influences. The street stretches about 1200 m. It four separate structures. Each tetrapylon consists of four was built in three stretches over the course of a century columns which are topped by a solid cornice (Fig. 19). starting from the 2nd century, but never finished. Columns are supported by plints at the corners of a The western portion was the first part to be built, it was square. It stood in a circular piazza marking the junction also the longest, stretching over 500 m (Butcher, 2003). of Central and Eastern sections of the colonnaded street The middle portion connected the western and eastern (Fig. 20). stretches. A Tetrakionion (Sec. 5.2) marked where the If one progresses south-eastward through the Eastern Central section met the Eastern section. Eastern section section of the Grand Colonnade he comes to the Arch of ended with the Triumphal Arch. The width of the street Triumph. The Arch of Triumph was built in the 3rd extends to 22.7 m at the widest places and diminishes to century, after the propylaeum of the Temple of Bel was 11.7 m at the narrowest parts (Barański, 1995). finished. It gave an uninterrupted line of sight towards The columns were topped with Corinthian capitals and the great temple (Fig. 21). mounted on Attic bases. Despite the character of the When the traveller reaches from Tetrakionion to the Arch columns, pertaining to the characteristics of ancient of Triumph his gaze is directed inevitably towards the Greek and Roman architectures, they also carry some gate of the temple which was probably the cause why indigenous features. A noteworthy example of these is the Arch of Triumph had been placed at this point. A the presence of brackets half-way up each column. They possible explanation of their origin might be the sacred were located to hold the statues of dignitaries. processions. The colonnaded street itself was the most significant of It is tempting to conjure the Eastern section as a the traits of Eastern tradition in Palmyra. This is processional road which started at Tetrakionion, because, in contrast to the ubiquitousness of the continued through the Arch of Triumph, and ended at the colonnaded streets in the Eastern provinces of the Temple of Bel. How the Arch of Triumph was connected Roman Empire, they were almost completely absent in to the temple is not known. This part of the Grand the Western provinces (Ball, 2014). What lied behind the Colonnade was never finished. colonnades were rows of shops. Dedicatory inscriptions were found on the columns. One of them was dedicated to Zenobia. Grand Colonnaded Street of Palmyra (Fig. 18) was arguably the most famous of the surviving colonnaded streets until it was destroyed in October 2015. It had given to Palmyra an architectural unity, a common thread that brought Eastern, Western and Central sections of the city together (Watkin, 2011).

Figure 19: Drawing of the Great Tetrakionion [in m] (After Wiegand, 1932)

Figure 18: Reconstructed view of Grand Colonnaded Street.

6.2. Triumphal Arch and Great Tetrakionion Grand Colonnaded Street changes direction at two locations in order to reach the Temple of Bel. One of these is punctuated by the Tetrakionion where Central section meets the Eastern section. The other is marked by the Arch of Triumph. Figure 20: Reconstructed view of the Great Tetrakionion.

Virtual Archaeology Review, 8(17): 20-30, 2017 26 REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE

not only the monuments but the entire city of Palmyra has been brought back from rubble, by using the techniques of virtual reality. The topographical model of the landscape was covered with the natural fauna and vegetation. Its “pleasant streams“ as described by Pliny were also added (Fig. 25). With the illumination effects of sun setting and the walls of the temples turning gold and pink in the desert light, Virtual Palmyra makes one feel a deep connection across the centuries to the city’s heyday when Queen Zenobia was walking through the Grand Colonnade to the Temple of Bel (Fig. 26). Figure 21: Triumphal Arch and the eastern end of the Grand Colonnaded Street. 8. Conclusions It is tempting to conjure the Eastern section as a Palmyra was not only historically significant, but it was processional road which started at Tetrakionion, also beautiful. The 150000 tourists who were visiting it continued through the Arch of Triumph, and ended at the every year used to be entranced by what they saw: The Temple of Bel. How the Arch of Triumph was connected ruins of an ancient city rose in splendour as an oasis in to the temple is not known. This part of the Grand the surrounding Syrian Desert. Regrettably, since the Colonnade was never finished. end of 2015 summer, the view that tourists used to savour no longer exists. Ancient Palmyra had crumbled 6.3. The Theatre into obscurity in the surrounding desert. The Theatre of Palmyra is another building which has a The defining feature of the devastation in Palmyra was totally western form. The theatre is a horseshoe-shaped the intentional targeting of the cultural and religious (Fig. 22) building with a cavea of 92 m in diameter and heritage. The assaults were directed at monumental an adjoining portico. It was surrounded by a ring-shaped structures which symbolised and contained material area of 104 x 82 m (Figs. 23 and 24). evidence of Palmyra's multi-cultural identity. Traditional methods of cultural heritage conservation are of little use when one is faced with such a deliberate targeting. After the city was recaptured, the scale of devastation was seen: The UNESCO WHS was reduced to rubble and almost integrated with the enclosing desert. With looting and destruction of the relics of the historical city, irreplaceable evidences of ancient life and societies were lost forever. 3D computer graphics, through technological innovations, offers an ability in ‘reconstructing the past’ beyond those originally imagined. Considering that the disappearance of heritage is an impoverishment of the intellectual wealth of all nations, the use of this capacity of 3D computer graphics in archaeology and cultural Figure 22: Scaled bird’s eye drawing of the Theatre of Palmyra heritage entices careful consideration. Starting from the [in m]. 90s (Forte and Siliotti, 1997) the use of 3D computer graphics in relation with archaeology and cultural This expansive area had been very likely to be used heritage has been a focus of attention for scholars in both by the visitors of the theatre and by the caravan multi-disciplinary fields. An abundance of publications dealers who were coming to market their merchandise at have emerged in the last two decades which have the Agora. This area was excavated in (1959-1962). It presented photo-realistic reconstructions of the past was buried by then under 5 m high sand. It was brought (Logothetis et al., 2016) and (Manferdini et al., 2016). to daylight only after several thousand tons of rubble and This stream continues to make contributions to the sand were removed especially in the western and north- creation of a new legacy in cultural heritage. western areas. With the advent of affordable 3D printers, a new window has been opened to a new horizon. This development 7. Holistic virtual representation of enables tangible reconstruction of lost edifices of cultural Palmyra heritage in full 3D. The application of the state-of-the-art 3D printing technologies to Virtual Palmyra project has The reconstructed individual models of the temples and already provided successful results (Fig. 27). other landmarks of Palmyra have been pieced together Visualisation opportunities that are increasing being with the 3D environment to form the virtual model of developed by technological progress have been opening Palmyra as a whole. new horizons for reminding the past existence of the lost This stage of visualisation was imperative for the treasures of our cultural heritage. As new technological reconstructed city to accurately represent its now creations stream on, we become more and more able to demolished counterpart in reality. It is at this stage that virtually recreate the ruined cities and flattened

Virtual Archaeology Review, 8(17): 20-30, 2017 27 DENKER, 2017

Figure 23: Reconstructed view of the theatre. Figure 24: Reconstructed close-up view of the stage.

Figure 25: “Pleasant streams“of Palmyra as described by Pliny.

Figure 26: Temple of Bel and the 3D environment.

Virtual Archaeology Review, 8(17): 20-30, 2017 28 REBUILDING PALMYRA VIRTUALLY: RECREATION OF ITS FORMER GLORY IN DIGITAL SPACE

monuments and their past. Reconstructions of the past with the advent of 3D computer graphics, high resolution rendering and 3D printing are increasingly produced and maintained in digital form, thus creating a new type of legacy: Digital cultural heritage. With this work Virtual Palmyra has been created as a new edition to our digital cultural heritage. Virtual Palmyra is one of the means for depriving the terrorists of the opportunity of deleting Palmyra from our collective memory.

Acknowledgements The author acknowledges with pride the contributions of his students Berk Gencan, Eren Can Yildirir, Mehmet Cicekler and Meltem Sezer in the creation of the 3D Figure 27: Temple of Bel after 3D printing. reconstructions of this project.

References Al-Mounes, M. (2016). Assessing the damage at Palmyra. Retrieved from https://gatesofnineveh.wordpress.com/ 2016/03/31/assessing-the-damage-at-palmyra/ Ball, W. (2016). Rome in the East: the transformation of an empire, 2nd Ed. London & New York: Routledge. Barański, M. (1995). The Great Colonnade of Palmyra reconsidered. ARAM Periodical, 7(1), 37–46. http://doi.org/10.2143/ARAM.7.1.2002216 Butcher, K. (2003). Roman Syria: And the Near East, Paul Getty Trust Publications. Campbell, G. (2007). "Palmyra", The grove encyclopedia of classical art and architecture. Oxford University Press. http://doi.org/10.1093/acref/9780195300826.001.0001 Cassas, L. F. (1900). Voyage pittoresque de la Syrie, de la Phoenicie, de la Palaestine et de la Basse Aegypte. Paris: Impr. de la République. Available at http://digi.ub.uni-heidelberg.de/diglit/cassas1800bd3/0001

Collart, P. (1970). Reconstruction du Thalamos du Temple de Baalshamîn a Palmyre. Revue Archéologique, Nouvelle Série, 2, 323–327 Denker, A. (2016). Virtual Palmyra: 3D reconstruction of lost reality of “the bride of the desert”. In J. L. Lerma & M. Cabrelles (Eds.), Proceedings of Arqueológica 2.0, 8th International Congress: Advanced 3D Documentation, modelling and reconstruction of cultural heritage objects, monuments and sites (pp. 318–320). Valencia: Editorial Politècnica de València. Evans, R. (1995). The projective cast: architecture and its three geometries. MIT Press. Forte, M., & Siliotti, A. (1997). Virtual archaeology: re-creating ancient worlds. London: Harry N Abrams B.V. O’Brien, J. (2015). Palmyra: ruins that inspired the architecture of power. In BBC News, Washington, 27 July 2015. Available at http://www.bbc.com/news/magazine-33659376 Gawlikowski, M. (1995). Bel of Palmyra. In M. Bomer-A. Lichtenberger-R. Raja (Eds.), Religious identities in the Levant from Alexander to Muhammed (pp. 247–255). Brepols Publishers.

Logothetis, S., & Stylianidis, E. (2016). BIM open source software (OSS) for the documentation of cultural heritage. Virtual Archaeology Review, 7(15), 28–35. http://dx.doi.org/10.4995/var.2016.5864 Manferdini, A. M., Gasperoni, S., Guidi, F., & Marchesi, M. (2016). Unveiling Damnatio Memoriae. The use of 3D digital technologies for the virtual reconstruction of archaeological finds and artefacts. Virtual Archaeology Review, 7(15), 9– 17. http://dx.doi.org/10.4995/var.2016.5871 Pliny the Elder, Natural History 5.88.1. Reuters (2016). Palmyra's dynamited Temple of Bel can be restored, de-miners use robots. Available at http://blogs.reuters.com/faithworld/2016/03/31/palmyras-dynamited-temple-of-bel-can-be-restored-de-miners-use- robots/ Schlumberger, D. (1971). Les quatre tribus de Palmyre. Syria, 48: 121–133. Schmidt-Colinet, A. (1995). Palmyra: Kulturbegegnug im Grenzbereich. Verlag Philip von Zabern.

Virtual Archaeology Review, 8(17): 20-30, 2017 29 DENKER, 2017

Seyrig, H. (1950). Palmyra and the East. The Journal of Roman Studies, 40(1&2), 1–7. http://doi.org/10.2307/298497

Seyrig, H., Amy, R., & Will, E. (1975). Le Temple de Bel à Palmyre, Vol. I, II. Album, Texte et Planches, Institut Français d'Archéologie de Beyrouth, Bibliothèque Archéologique et Historique. T. 83. Paris: Librairie orientaliste Paul. Veyne, P., (2015). Palmyra Requiem für Eine Stadt, Verlag C.H. Beck oHG, München. Watkin, D. (2011). A History of Western Architecture. Laurence King Publishing 5th edition. Wiegand, T. (1932). Palmyra Ergebnisse der Expeditionen von 1902 und 1917. Text-und Tafelbd. Available at http://digi.ub.uni-heidelberg.de/diglit/wiegand1932ga Wood, R., (1753). The ruins of Palmyra, otherwise Tedmor, in the desert. Available at http://digi.ub.uni- heidelberg.de/diglit/wood1753 Segal, A. (2013).Temples and Sanctuaries in the Roman East, 95-99. Oxford: Oxbow Books.

Virtual Archaeology Review, 8(17): 20-30, 2017 30

Virtual Archaeology Review, 8(17): 31-41, 2017 http://dx.doi.org/10.4995/var.2017.6557 © UPV, SEAV, 2015

Received: September 6, 2016 Accepted: April 13, 2017

RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO LA RECREACIÓN DE UNA ESCENA URBANA MEDIEVAL CON PERSONAJES INTELIGENTES: PASOS PARA CREAR EL ESCENARIO COMPLETO Ana Paula Cláudioa,*, Maria Beatriz Carmoa, Alexandre Antonio de Carvalhoa, Willian Xaviera, Rui Filipe Antunesa,b

a BioISI - Biosystems & Integrative Sciences Institute, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal. [email protected]; [email protected]; [email protected]; [email protected]; [email protected] b MIRALab, University of Geneva, Switzerland

Abstract: From historical advice to 3D modeling and programming, the process of reconstructing cultural heritage sites populated with virtual inhabitants is lengthy and expensive, and it requires a large set of skills and tools. These constraints make it increasingly difficult, however not unattainable, for small archaeological sites to build their own simulations. In this article, we describe our attempt to minimize this scenario. We describe a framework that makes use of free tools or campus licenses and integrates the curricular work of students in academia. We present the details of methods and tools used in the pipeline of the construction of the virtual simulation of the medieval village of Mértola in the south of Portugal. We report on: a) the development of a lightweight model of the village, including houses and terrain, and b) its integration in a game engine in order to c) include a virtual population of autonomous inhabitants in a simulation running in real-time. Key words: virtual archaeology, digital archaeology, cyber-archaeology, cultural heritage, 3D reconstruction, 3D data interoperability.

Resumen: El proceso de reconstrucción en 3D de lugares históricos poblados con habitantes virtuales es complejo, y requiere de un gran conjunto de habilidades y herramientas. Estos procesos incluyen el asesoramiento histórico, el modelado 3D, o la programación y estas limitaciones hacen que sea muy difícil construir sus propias simulaciones en pequeños equipos arqueológicos. En este artículo, describimos nuestro intento de minimizar este escenario. Se describe cómo integramos el trabajo curricular de los estudiantes en la academia, haciendo uso de herramientas libres o licencias de campus. Se presentan los detalles de los métodos y de las herramientas utilizadas en el processo de construcción de la simulación virtual del pueblo medieval de Mértola, en el sur de Portugal. Se presenta: a) el desarrollo de los modelos ligeros 3D en el pueblo, incluyendo las habitaciones y el terreno, y b) su integración en un motor de juego con el fin de c) incluir la población virtual de habitantes autónomos, en una simulación que pueda ser ejecutada en tiempo real. Palabras clave: arqueología virtual, arqueología digital, ciber-arqueología, patrimonio cultural, documentación, reconstrucción 3D.

1. Introduction architecture and culture of the ancient site of Mértola. This input included information about the civil housing Amongst the challenges of creating a 3D simulation of and military structures in the urban landscape, such as ancient sites are the costs associated with these types of the defensive wall and the watchtower. Based on this simulations. One of our goals was to address this issue information, students’ activities were then organized by by building a pipeline of tools and methods available for teachers in such a way as to include tasks that would free in an educational setting or with campus licenses, allow the modeling of these features of the urban scene and basing it on volunteer work. This work resulted from (Figure 1). a multidisciplinary team of volunteers, including archeologists from the Campo Arqueológico de Mértola, Initially, three groups of students used manual modeling and students and teachers from Faculdade de Ciências to reproduce architectural models of the old village. One da Universidade de Lisboa working as part of their group modeled the iconic watchtower on the riverbank of curricular activities. the Guadiana in AutoCAD (AutoCAD, 2016). Another group modeled a sample of houses using AutoCAD and The former group of participants has provided useful Blender (Blender, 2016) which were later textured using insights and historically relevant guidelines about the 3ds Max (3ds Max, 2016). Finally, a third group

*Corresponding author: Ana Paula Cláudio, [email protected] 31

CLÁUDIO et al., 2017

elements of the scenario with animated virtual characters to obtain an interactive application. The main contribution of this paper is the identification of a set of techniques and procedures that can be used to attain the final interactive environment recreating the inhabited ancient village. We intentionally give some detail with the aim of easing the effort of producing a similar result for other case studies, targeting other places and epochs. Figure 2 illustrates an overview of the village of Mértola, populated with some virtual humans.

(a)

Figure 2: Detail of the model inside the defensive walls (b) populated with virtual characters.

Figure 1: Overview of the village inside the defensive walls: This paper is organized as follows: Section 2 present state (a) and our model (b). summarizes related work, Section 3 explains the setup of the virtual urban environment and Section 4 describes produced an agglomerate of fifteen houses of the the integration of virtual intelligent characters in this Alcáçova (citadel) in 3ds Max. environment. Finally, Section 5 discusses the results of The following step to generate the combined landscape the work, while Section 6 sums up the main goals departed from obtaining a model of Mértola’s terrain and achieved. placing these architectural models on top of it. This step was performed using CityEngine (CityEngine, 2016). 2. Related work Taking advantage of a CityEngine functionality called CGA (Computer Generated Architecture) shape In recent years, there has been widespread use of virtual grammar, the virtual landscape was enriched with environments that reproduce ancient civilizations streets, vegetation, defensive walls and a new house anchored in credible historical information. Without being model that was built combining polygonal modeling with exhaustive, we refer in the next paragraphs some rules. valuable works in this area. The first group of references concerns approaches that recreate ancient virtual Finally, the complete model was exported to a format architectural scenarios. Then, we direct attention to compatible with Unity (Unity, 2016), a software tool used works that include intelligent virtual humans. Finally, we to develop videogames. Then, virtual characters were look at some other examples that incorporate intelligent included in the scenario, the physical appearances of crowds. which are faithful to the epoch we intend to recreate. These characters exhibit simple, intelligent behaviors The methods adopted to recreate ancient virtual that in the future will be improved upon by algorithms scenarios from the architectural point of view may be inspired in artificial life (Antunes & Thalmann, 2016a; similar to the general approaches for urban generation Antunes & Thalmann, 2016b). (see Musialski et al., 2013, for a thorough survey). Concerning virtual recreation of historical places We obtained the models of the characters and their considerably damaged or that no longer exist, the basic animations from specialized software companies processes used are predominantly based on manual or like Daz 3D (Daz3d, 2016) and Mixamo Fuse (Mixamo procedural modeling. Fuse, 2016). Their spatial movements were accomplished by means of the physics engine and For instance, in the recreation of “La Venta”, the major algorithms incorporated in Unity; some scripts were used regional center of Olmec culture during the period 800- to tune movements and behaviors (e.g. character 400 BC in present-day Mexico, Gillespie & Volk (2014) following another character; donkey following a man). used manual modeling starting with plans mainly Our approach explores the interoperability of 3D produced from the archaeological excavations. Also software tools to combine the models of different resorting to archaeological excavations that had been

Virtual Archaeology Review, 8(17): 31-41, 2017 32

RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO made in Zaragoza, Spain, Basterra & Pintado (2015) Some reconstitutions elevate the challenge by including produced a virtual model of the “Forum Renascens”, the virtual populations with intelligent behaviors. For ancient public square of the Roman town, a place which instance, in the reconstruction of the agora of Argos, is still in process of research. The recreated 3D model Greece, Heigeas et al. (2003) created a crowd resorting was produced by manual modeling and constitutes, as to a physically-based particle model. Another example is the authors emphasize, a working piece for the team of Ciechomski et al. (2004), who created an audience of researchers that studies the historical place and that Intelligent virtual characters (usually called agents in includes architects, archeologists and historians. artificial intelligence) in the ancient Roman Odeon of Aphrodisias. With the aim of studying the ergonomics of Manual modeling can also be used to complete initial th the Roman Colosseum, Gutierrez et al. (2005), drafts. Carpetudo & Lopes (2015) recreated a 16 simulated a crowd of virtual agents whose movements century church in Montemor-o-Novo, Portugal, using are spatially uncertain. photogrammetry with images obtained with a drone. They produced an initial model of the contemporary More recently, Bogdanovych et al. (2010) deployed ruins and then manually completed the church based on intelligent agents in the online virtual world Second Life its representations in ancient paintings. (SecondLife, 2016) in a simulation of Uruk, an ancient Sumerian city. In the project “Rome Reborn,” (Frischer et al., 2008; Dylla et al., 2009) both manual and procedural modeling Building upon this last work, Trescak et al. (2014) were used to depict the evolution of the city of Rome presented an approach where objects have annotated from the late Bronze era until the 6th century AD. emotional responses. The agents in the scene, in the Procedural methods are also used by Rodrigues et al. course of their actions, choose actions and objects that (2014), who propose a grammar for generating virtual better fit their personality parameters. So, depending on heritage traversable house models. They demonstrated the particular configurations, certain aspects of one their method with the creation of several Roman houses individual might differ widely from those observed in from the heritage site of Conímbriga, Portugal. Another others, for instance, the level of tolerated proximity or approach consists in the parametric associative walking speed. geometry design tools used by Rua & Gil (2014) to model the 14 km length of the Lisbon Aqueduct. The work we describe in this paper continues this line of research and is part of a larger project called "BIHC - Fita, Besuievsky & Patow (2017) present a set of Bio-Inspired Human Crowds" whose partial objectives techniques for virtual reconstruction that merge are the development of a set of tools for the animation of procedural modeling with structural analysis methods. 3D reconstructions of cultural heritage sites1. This paper The aim of these techniques is to provide models that describes a combined effort to work under a low-budget are realistic from two perspectives: visual accuracy and pipeline to make a 3D simulation of an ancient village. structural feasibility. The faithfulness of the model is Our case study is the medieval village of Mértola in the crucial to understand how structural elements such as south of Portugal, and the work involves a bricks, stone blocks or vaults, were combined in the multidisciplinary team of volunteers comprising construction of the building; this constitutes valuable archeologists in Mértola, and students and teachers from information for preservation and maintenance actions. Faculdade de Ciências da Universidade de Lisboa These authors describe, as a case-study, a model of a working as part of their curricular activities. Romanesque church. Besides the architectural structures of the sites, some 3. Creating the virtual scenario authors explore other facets of the simulations, The creation of the virtual scenario combines several implementing approaches commonly used in elements in the same environment: terrain, streets, videogames (see Anderson et al., 2010, for a review houses, fortifications (the watchtower and the defensive about serious games for cultural heritage). Three recent wall) and vegetation. Figure 3 highlights the main examples are provided by Capurro, Nollet & Pletinckx components used in the pipeline. We have applied the (2014), Saldaña (2015) and Kennedy et al. (2013). geographic processing capabilities of the ArcGIS Capurro et al. (2014) conceived a virtual model platform (version 10.3.1; ArcGIS, 2016) to import the (including architectural decoration and furnishing) of the th model of the terrain, and AutoCAD (version 2016), 13 century Saint Saviour abbey in Ename, Belgium, Blender (version 2.73) and 3ds Max (version 2015) to with the objective of releasing an educational game for model the architectural features. The resulting models young visitors of the Provincial Heritage Centre, nearby were then exported to CityEngine (version 2014.0), the archaeological site of the abbey. The players will where more elements were added mainly by procedural walk inside the virtual abbey, discover the daily life of modeling. Afterwards, the complete virtual environment monks, while overlooking the real archaeological site. was exported to Unity (version 5.3.3) in FBX format, and Saldaña (2015) presented a suite of procedural rules the intelligent virtual characters were integrated in the that generate buildings for modeling Greek and Roman final simulation. cities. In the work, the observer is represented as an avatar in the virtual environment and is able to navigate and explore the scenario. Kennedy et al. (2013) used 3.1. Terrain OpenSim to recreate St. Andrews Cathedral in Scotland Three main steps were carried out to obtain the model of th during the 14 century. They populated the scenario with the terrain. The first, Input Datasets, was required to virtual characters. Additionally, the user controls the acquire the topography of the terrain in the area of movements of an avatar, which can interact with several elements of the environment. 1 http://bihc-fcul.weebly.com/

Virtual Archaeology Review, 8(17): 31-41, 2017 33

CLÁUDIO et al., 2017

guidelines described in a DGT document2, namely, selecting Window Catalog/Interoperability Connections, choosing Concelhos3_2015 and finally our target municipality, Mértola, whose limits appear in the interface. A new shapefile, called points_of_control, was created selecting the feature type (in this case the type point) and the corresponding coordinate system (in our case-study, WGS 1984 UTM Zone 29N). After that, the data was transferred. This shapefile was dragged and dropped into the interface of ArcMap where a new layer was defined. Then we edited it to configure the latitude and longitude of the four corners of the region of interest obtained from OpenTopography services. By the end of this process, the measurement units were converted to meters.

3.1.2. Step two – Image generation Figure 3: Pipeline overview. With ArcCatalog, the shapefile created in the previous interest that our virtual model reproduces. The second step (points_of_control) was exported to the KML format. step, Image Generation, was necessary to get a texture The resulting file was imported into Google Earth Pro to for the corresponding area. The third, Data and Image achieve a high resolution image (in JPG format) that Processing, combined the results of the two previous involves the region of interest with the four delimiting steps and prepared the files to be transferred to control points marked with glyphs. CityEngine. Figure 4 illustrates this sequence, which is detailed next. 3.1.3. Step three - Data and image processing In ArcMap, we similarly created a new layer to import the image produced in the previous step. Then, using the four coordinates marked in the image and the shapefile points_of_control generated earlier, we georeferenced the image using the Georeferencing tool. This process was repeated for each point. Previously, we disabled the option Auto Adjust in View Link Table. Furthermore, we also cropped the image using the coordinates defined by the mentioned four points so that only the inside area remained (mosaic to new raster tool). These two operations assured that the image fit as a texture for the area of interest. Next, the previously mentioned GeoTIFF file was imported to ArcMap in order to produce the map of elevation of the terrain Figure 4: Steps to obtain the textured terrain. (height map). This map was generated using the ArcToolbox functionality Spatial Analyst Tools. 3.1.1. Step one - Input datasets Finally, this procedure was completed by importing and The whole process started with the selection of the combining the produced height map and texture to geographical region of interest where the 3D scene CityEngine. This was done by creating a new project in should be located. In the Google Earth interface, we CityEngine and then importing the mentioned files under identified the region consisting of the ancient village of the menu option Layer/New MapLayer (Fig. 5). Mértola as an area of 1 km2. We then obtained the village’s upper and lower limits expressed in terms of 3.2. Streets and shapes longitude and latitude. Next, we used OpenTopography (OpenTopography, 2016), a service that offers raster The CityEngine feature Grow Street was used to datasets with different granularities for digital terrain generate the streets as well as the blocks in between models (DTM). The values of latitude and longitude these streets where the buildings were placed. This established the four corners of the region of interest. We feature uses the heightmap file and an obstacle file that then generated a corresponding image in GeoTIFF in our case study prevents the appearance of streets on format containing the information about the elevation of the riverbeds. the terrain. This image file was then processed with the The obstacle file (Figure 6) was obtained by editing the ArcGIS platform to be combined with a texture using its georeferenced image on the left side of Figure 5. The tools: ArcMap, ArcCatalog and ArcToolBox. streets were textured to simulate gravel using rules. The ArcCatalog application was used to access information provided by the Portuguese entity Direção Geral do Território (DGT). The first step of this process was the configuration of this service following the 2mapas.dgterritorio.pt/Openviewer/cos2007/Guia_de_apoio_utili zacao_WFS_ArcGIS.pdf 3 Concelho is the Portuguese word for municipality.

Virtual Archaeology Review, 8(17): 31-41, 2017 34

RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO

(a) (b) Figure 5: Terrain: a) Georeferenced image and b) digital Figure 7: Model of the houses in the Alcáçova (3ds Max). The elevation model. houses highlighted have been modeled with 3 LoD

The models were finally imported into CityEngine as static models and were placed on the terrain using the option Align static models. Afterwards, the watchtower model was converted into shapes to be textured resorting to procedural rules. Figure 8 summarizes the two approaches to use the imported models in the CityEngine context.

Figure 6: Image in the obstacle file.

3.3. Houses and fortifications The virtual model of the village comprises houses, the watchtower and the defensive walls with its towers.

3.3.1. Houses and watchtower We resorted to manual modeling to reproduce three sets of constructions. The first set contains the complete structure of the watchtower; the second set contains three models of characteristic houses of Mértola and, based on guidelines provided by the archeologists, the Figure 8: The pipeline inside CityEngine. third set comprises the houses of the Alcáçova. The archaeological remains of the watchtower and the The watchtower was modeled in AutoCAD. The models houses of the Alcáçova are still easily identifiable of the second set were also created with AutoCAD nowadays. As such, we were able to place the however, for curricular reasons of the students involved, corresponding models in their exact locations on the some of them were also created with Blender. Finally, corresponding virtual terrain of Mértola. The models of the houses of the Alcáçova were modeled in 3ds Max. the Alcáçova houses follow the real plants; the height of Some of the houses have variations that allow the the walls and the characteristcs of the roofs were integration of different levels of detail (LoD): a) a provided by the archeologists of Campo Arqueológico de lightweight version with a geometry of a reduced number Mértola. Concerning the watchtower model, its of polygons, to be visible at a distance; b) a second measures and architectonic details were also provided version, where houses only have the exterior walls but by the same archeologists. have a higher definition with detailed roofs and doors, to be seen at close range; c) a third version, also in high The second set of houses mentioned earlier were used definition, with the interior walls and props allowing a as templates to create house clusters in general areas of visit using a first person point of view. Figure 7 gives a the village where we did not have much available combined overview of the three sets of houses side by information. To cover these unexcavated areas, we also side, which give no telltale signs of being different in used some house models generated in CityEngine by their making. polygonal modeling combined with procedural rules. We have also created narrow streets similar to the ones in All house models were textured in 3ds Max. Then, when the Alcáçova area. exporting in the OBJ format, two files are created: one with the geometry (extension .OBJ) and another with the 3.3.2. Defensive wall texture (extension .MTL). Although it is possible to export files directly from Blender to CityEngine, we also The approximate model of the defensive wall was based chose to apply the textures in 3ds Max for these models on information gathered from different sources. The because this software organizes the textures in such a location of the wall on the terrain follows a 2D map way that eases the import process.

Virtual Archaeology Review, 8(17): 31-41, 2017 35

CLÁUDIO et al., 2017

obtained in Martínez et al. (2009), while the location and was positioned, and Figure 10b shows the final result. shape of the its towers were inspired by Macias (1996b; Figure 11 displays a detail of a tower integrated in the 3 towers near the Alcáçova) and on hand drawings defensive wall. To apply textures to the defensive wall, made in situ by Duarte Darmas in the 16th century (this its towers and to the watchtower we used procedural author identified 13 towers around the defensive wall). rules. Although these drawings have been depicted in an epoch that is posterior to the one we want to recreate, we used them as they present many valuable details. The first step of this process utilized polygonal modeling in CityEngine to produce five different models of defensive towers to form the modular basis of the defensive wall (Figure 9). These models were exported in the .OBJ format to be used as assets by CityEngine in the code rule that generates the final walls.

Figure 9: The 5 models of defensive towers produced by polygonal modeling in CityEngine. Figure 11: Detail of a tower inserted in the wall. The second step to generate the complete defensive wall employed in the Grow Street feature, followed by 3.4. Combining the 3D elements the necessary manual adjustment to approximate the To combine the terrain and the models described in layout of the streets as much as possible to the contour Section 3.3, we have defined layers for: a) the terrain, b) of the defensive wall, as depicted in a) Martínez (2009), the static models; and c) the shapes. The shape layers b) Macias (1996b) and c) Duarte Darmas (1509), as well include models generated by grammar rules. These as d) the current images obtained by Google Earth. include buildings and streets, as well as landscape These streets worked as the basis for lifting the wall. A creation with tree models from a vegetation library4. We rule was used to split the street in two halves and two used alignment features to combine the 3D models with distinct rules were applied to each half: on the external the terrain: Align graphic to terrain, Align terrain to shape half, a rule creates a wall with merlons, and on the inner and Align static model. half another rule generates a lower interior wall where, for instance, the sentinels could stand while watching the The complete virtual 3D model generated in CityEngine surroundings. The towers were inserted in some street was exported to Unity using the FBX format following the intersection nodes. Their models fit the ones depicted in instructions provided5. Moreover, we disabled the option a) Macias (1996b) and b) Duarte Darmas (1509). Figure Create Shape Group to reduce the number of polygons 10a highlights a street intersection node where a tower exported, a necessary concern to ensure a lightweight model that supports navigation in real-time. Also considering this constraint, some elements, such as the houses, were modeled with different LoD, as previously mentioned, and set in distinct layers before being exported. This way, the Unity game engine can optimize rendering processes, adjusting the complexity of the mesh according to the distance of the camera. Models that are away from the camera are displayed with fewer polygons while the closer ones are rendered with more polygons to attain a more realistic effect. For all the other models, we also had the concern of keeping the number

4 www.arcgis.com/apps/CEWebViewer/viewer.html?3dWeb (a) (b) Scene=6015f7d48dff4b3084de76bcf22c5bca Figure 10: Streets used to create the wall with a node on the 5 “How to: Export Textured Models from CityEngine into Unity”, intersection (a) and the final result with a tower on the node (b). http://support.esri.com/technical-article/000012449

Virtual Archaeology Review, 8(17): 31-41, 2017 36

RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO of polygons low in a way that would not compromise the To ensure the credibility of the movements exhibited by credibility of the model. the characters, two factors are critical: the quality of the initial animations and the reproduction of the laws of Finally, part of the housing complex in the interior of the physics. Unity’s physics engine uses colliders to deal walls was reconstructed based on direct evidence from with collisions amongst objects. As such, to guarantee the excavations. However, there was also a large that characters move on the ground and do not exhibit undocumented section of the village for which we had to unnatural movements, like trespassing walls, we need to extrapolate and interpret based on the available define colliders for every mesh corresponding to material. After the first results, and based on the elements in the scenario that may be in the way of a feedback from the archeologists, these sections that did moving character, such as characters, houses, walls or not follow archeological evidence, had to be redone: trees. Similarly, the terrain also incorporates a collider to a) to narrow the width of the streets, and b) to augment ensure the virtual characters will move on its surface, the urban density. The final model, depicted from two preventing the force of gravity (implemented by the different perspectives in Figure 2 and Figure 12, reflects physics engine) from making them fall vertically and these adjustments. disappear from the scene.

5. Discussion Building a simulation, such as the one described in this paper, raises organizational and implementational questions. In terms of organization, when we compare the present work with others of the kind, such as Pompeii (Maim et al., 2007), the nature of the present work carries inherent risks and increased difficulties since this project is based on construction by volunteers and is built using tools available with free or campus licensing. This has generated a number of extra hurdles to its successful completion. We had to coordinate multiple tasks involving people working on an irregular basis and consequently we all had to be very flexible in Figure 12: Overview of the walls, with the village in the our approaches, and to integrate and articulate the background and the watchtower on the right. progress of the related activities. One example is the interdependence of a) retrieving the 4. Inserting the animated characters historical information, b) modeling the architecture of a The methodology to incorporate animated characters in specific building; c) texturing the resulting model; the virtual environment consisted of two major steps: d) validating the historical accuracy of the resulting 1) the acquisition and animation of the models; and model; e) creating LoD for the object; f) incorporating the 2) integration into the environment. Such integration was final model into CityEngine and Unity. As these activities expanded providing intelligent behavior to the had short time slots to be completed – normally part of a characters, for instance, making them capable of semester – and students had no previous experience in choosing the shortest path between two points. such technical tasks, the difficulties increased significantly with a negative impact in the quality and The population present in the final scene includes men, volume of work. Yet, with several adjustments on our women, children and donkeys. Some preparatory work part, we are all very pleased with the results achieved was required to configure the respective models. The since we were able to successfully deliver the simulation initial models of the virtual humans were obtained from with our limited resources. A visualization of a lighter Mixamo Fuse (v. 1.0), and later these rigged models version of the raw 3D models can be accessed at were exported to 3ds Max where they were adequately http://www.arcgis.com/apps/CEWebViewer/viewer.html? dressed with garments from the recreated epoch. The 3dWebScene=442744f3d8964b6eb921b2e0adef4b8a. number of vertices was reduced (MultiRes functionality) The final simulation is available using Unity and they were exported to FBX with the option Embed Webplayer at the following link: http://bihc- Media. Then, Mixamo was used to automatically adjust fcul.weebly.com/myrtula.html. skeleton definitions and skinning. This tool was also used to provide a set of basic movement animations A second thread of practical questions relates to the (e.g., walk, run, sit on a bench, climb stairs). technical complexity of bringing together architecture and vegetation in an urban layout enriched with virtual Animal characters, such as the donkey, were obtained characters in real-time rendering. The Unity game from Arteria3D (Arteria3D, 2016) already preprocessed engine provides an efficient mechanism of occlusion and ready to be integrated into Unity. culling, which minimizes part of the problem. The basic animations of the characters were combined Additionally, to reduce the number of simultaneous with scripts included in Unity that implement simple polygons being rendered per frame, we also had to use artificial intelligence algorithms, such as the A* algorithm LoD for the objects. As the complexity and detail of the for choosing the shortest path between two points or models decrease with the distance to the camera, the “follow target” for making one character follow another. associated loss of information is negligible, thus Figure 2 and Figure13 illustrate situations involving preserving the accuracy of the model. Finally, we also virtual characters moving in the scenario according to maximized the number of shared materials to minimize these simple algorithms. the number of GPU calls.

Virtual Archaeology Review, 8(17): 31-41, 2017 37

CLÁUDIO et al., 2017

(a) (b) Figure 13: Virtual characters populating the model: a) a group outside the defensive walls and b) two characters interacting with each other.

(a)

(b) Figure14: a) The architectural features built based on the archaeological evidence: the houses in the Alcáçova and the watchtower (inside the red squares); b) The final composition, where these structures appear combined with our own interpretation of the provided information.

The set of tools that we used support compatible formats using the same geographical coordinates, the for data import and export, which provides data corresponding aerial image. This image superimposed interoperability among them. This fact was crucial to take over the 3D model of the terrain made possible the advantage of specific features of each tool, and to define correct positioning of the models whose ruins are still a pipeline to accomplish the final goal of the project. In visible today (Alcáçova and the watching tower). particular, terrain generation required the combination of data from different sources. We were able to obtain an Despite the arguable merits of the resulting simulation, accurate georeferenced 3D model of the terrain and, it is noteworthy to mention that the project is still a work

Virtual Archaeology Review, 8(17): 31-41, 2017 38

RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO in progress, and we are still gathering relevant have identified and discussed the issues that were archaeological information necessary for the historical raised during the process of construction. accuracy of the simulation. First and foremost, not all of the iconic buildings of the village appear in the We generated an environment that combines 3D models simulation, such as the mosque. This situation might created by both manual and procedural modeling with a raise incorrect, or rather incomplete, perceptions of the realistic model of the terrain. Moreover, this 3D ancient site to the unaware visitor. We plan to mitigate environment scene is inhabited by intelligent virtual this with further work on the 3D modeling in the characters. The building elements (houses, defensive following version. Another problematic fact is that we walls and watchtower) and the appearance of the have combined buildings appearing in their exact characters are faithful to the epoch we intend to recreate. historical locations next to other ones that were Currently, the virtual model of the medieval village of inserted to fill in gaps for the time being (Fig. 14). Mértola is being improved upon. In the previous section, Besides using different texturing properties to we mentioned the need to improve the current simulation distinguish the former from the later, we also plan to fix through the inclusion of some of the iconic buildings. this situation in the near future by allowing the visitor to Concerning the virtual inhabitants, algorithms inspired in switch the view of these buildings on and off as to artificial life are also being developed to obtain a distinguish the historically accurate buildings, and we generative model for heterogeneity and spontaneity in will make adjustments to the simulation as we get behavioral animation of small, medium and large further archaeological information from our partners at gatherings of virtual characters. Campo Arqueológico de Mértola. Overall, we have set ourselves the challenge of reconstructing the medieval village of Mértola. This is 6. Conclusions and future work still an ongoing process, but so far our experience with We have described the pipeline used in our ongoing this project provides a useful and practical example of work of constructing a simulation of the medieval village the use of low-budget tools and means to build this of Mértola. Our main contribution is the identification of a type of simulation. This fact is of great help to other set of tools and stages that can be used to obtain the small teams similarly constrained by their budgets, but final interactive environment recreating the inhabited interested in developing this type of educational tool. ancient village. In this article, we provide a detailed description of the procedures used in creating the Acknowledgements terrain, the urban features and the characters used in the simulation. This project was funded by the Horizon 2020: EU Framework Programme for Research and Innovation We began this article by identifying our motivations for under Marie Skłodowska-Curie grant agreement No. using a low-budget process that includes free licensing 655226 (fellowship to Dr. Rui Filipe Antunes). We tools or academic licenses. We have also referred to the acknowledge the support of BioISI R&D unit requirements of the simulation being accessed by an (UID/MULTI/04046/2013), Portugal, the information online audience a) in real-time and b) in immersive mode provided by the team of Campo Arqueológico de using the first person point of view, and the consequent Mértola and the students that produced the models and need for using lightweight 3D models. Furthermore, we scenarios. Proofreading by Piki Productions.

References 3ds Max (2016). Retrieved in January 2016 from: http://www.autodesk.com/products/3ds-max/overview Anderson, E.F., Mcloughlin, L., Liarokapis, F., Peters, C., Petridis, P., & Freitas, S. (2010). Developing serious games for cultural heritage: a state-of-the-art review. Virtual Reality, 14(4), 255–275. http://dx.doi.org/10.1007/s10055-010- 0177-3 Antunes, R. F., & Magnenat-Thalmann, N. (2016a). Human crowd simulation: what can we learn from alife?”. In Artificial Life Conference 2016 (pp. 38–45). New York, NY, USA: MIT Press. http://dx.doi.org/10.7551/978-0-262-33936-0- ch012

Antunes, R. F., & Magnenat-Thalmann, N. (2016b). Bio-inspired virtual populations: adaptive behavior with affective feedback. In 29th International Conference on Computer Animation and Social Agents (pp. 101–110). New York, NY, USA: ACM. https://doi.org/10.1145/2915926.2915929 ArcGIS (2016). Retrieved in January 2016 from http://desktop.arcgis.com Arteria3D (2016). Retrieved in January 2016 from http://arteria3d.myshopify.com AutoCAD (2016). Retrieved in January 2016 from http://www.autodesk.com

Basterra, P. S., & Pintado, J. A. (2015). Forum Renascens (Los Bañales de Uncastillo, Zaragoza): Archaeology of Architecture of the Roman forum in the service of the dissemination through the Virtual Archaeology. Virtual Archaeology Review, 6(12), 109–121. http://dx.doi.org/10.4995/var.2015.4166 Blender (2016). Retrieved in January 2016 from http:www.blender.org

Virtual Archaeology Review, 8(17): 31-41, 2017 39

CLÁUDIO et al., 2017

Bogdanovych, A., Rodriguez-Aguilar, J. A., Simoff, S., & Cohen, A. (2010). Authentic interactive reenactment of cultural heritage with 3D virtual worlds and artificial intelligence. Applied Artificial Intelligence, 24(6), 617–647. http://dx.doi.org/ 10.1080/08839514.2010.492172 Capurro, C., Nollet, D., & Pletinckx, D. (2014). Reconstruction of the interior of the Saint Salvator abbey of Ename around 1290. Virtual Archaeology Review, 5(11), 34–41. http://dx.doi.org/10.4995/var.2014.4174 Carpetudo, C., & Lopes, G. (2015). The old parish church of Montemor-o-Novo in the 16th century-a manuelino example and its virtual reconstruction. In Digital Heritage 2015, (pp. 729–730). http://dx.doi.org/10.1109/DigitalHeritage.2015.7419614 Ciechomski, P. H., Ulicny, B., Cetre, R., & Thalmann, D. (2004). A case study of a virtual audience in a reconstruction of an ancient Roman Odeon in Aphrodisias. In 5th International Symposium on Virtual Reality, Archaeology and Cultural Heritage (VAST 2004), (pp. 47–56). http://dx.doi.org/10.2312/VAST/VAST04/009-017 CityEngine (2016). Retrieved in January 2016 from http://www.esri.com Darmas, D. (1509). Livro das fortalezas situadas no extremo de Portugal e Castela por Duarte de Armas, Escudeiro da Casa do Rei D. Manuel I. Retrieved from http://dx.doi.org/digitarq.arquivos.pt/viewer?id=3909707 Daz3d (2016). Retrieved in January 2016 from http://www.daz3d.com Dylla, K., Frischer, B., Muller, P., Ulmer, A., & Haegler, S. (2009). Rome Reborn 2.0: A case study of virtual city reconstruction using procedural modeling techniques. In 37th CAA Conference 2009, (pp. 62–66). Fita, J., Besuievsky, G., & Patow, G. (2017). A perspective on procedural modeling based on structural analysis. Virtual Archaeology Review, 8(16), 44–50. http://doi.org/10.4995/var.2017.5765 Frischer B., Abernathy D., Guidi G., Myers J., Thibodeau C., Salvemini A., Muller P., Hofstee P., & Minor, B. (2008). Rome Reborn. In ACM SIGGRAPH new tech demos (SIGGRAPH'08), Article 34. Gillespie, S. D. , & Volk, M.(2014). A 3D model of Complex A, La Venta, México. Digital applications in Archaeology and Cultural Heritage, 1(3), 72–81. http://dx.doi.org/10.1016/j.daach.2014.06.001 Gutierrez, D., Fricher, B., Cerezo, E., Sobreviela, E., & Gomez, A. (2005). Virtual crowds in a digital Colosseum. In Virtual Retrospect 2005, (pp. 82–87). http://dx.doi.org/10.1016/j.culher.2007.01.007

Heigeas, L., Lucianiz, A., Thollotx, J., & Castagnéz, N. (2013). A physically-based particle model of emergent crowd behaviors. In Graphikon’03, (pp. 5931–5939). http://artis.imag.fr/Publications/2003/HLTC03/crowd_Graphicon.pdf Kennedy, S., Fawcett, R., Miller, A., Dow, L., Sweetman, R., Field, A., Campbell, A., Oliver, I., McCaffery, J., & Allison, C. (2013). Exploring Canons and Cathedrals with Open Virtual Worlds: The recreation of St Andrews Cathedral, St Andrews Day, 1318. In International Congress on Digital Heritage, Vol. 2, (pp. 273–280). http://dx.doi.org/10.1109/DigitalHeritage.2013.6744764 Macias, S. (1996a). Mértola Islâmica: Estudo Histórico-Arqueológico do Bairro da Alcáçova, Séculos XII-XIII. Campo Arqueológico de Mértola. Macias, S. (1996b). Islamic quarter. Accessed on 30-05-2016: http://www.discoverislamicart.org/ database_item.php?id=monument;ISL;pt;Mon01;9;en Martínez, S., Lopes, V., Torres, C., Palma, M. F., & Macias, S. (2009). XXII Mértola Islâmica. A madina e o arrabalde. Campo Arqueológico de Mértola. Accessed on 13-06-2016: https://comum.rcaap.pt/handle/10400.26/2130

Mixamo Fuse (2016). Retrieved in January 2016 from http://www.mixamo.com/fuse Musialski, P., Wonka, P., Aliaga, D. G., Wimmer, M., Gool, L. V., & Purgathofer, W. (2013). A survey of urban reconstruction. Computer Graphics Forum, 32(6), 146–177. http://dx.doi.org/10.1111/cgf.12077 Muller, P., Wonka, P., Haegler, S., Ulmer, A., & Gool, L. (2006). Procedural modeling of buildings. ACM Transactions On Graphics, 25(3), 614–623. http://dx.doi.org/10.1145/1179352.1141931 Maim, J.,Haegler, S.;Yersin, B.,Muller, P.,Thalmann, D., & Van Gool, L. (2007). Populating ancient Pompeii with crowds of virtual humans. In 8th International Symposium on Virtual Reality, Archaeology and Cultural Heritage – VAST2007 (pp. 26-–30). OpenTopography (2016). Retrieved in January 2016 from http://opentopo.sdsc.edu/datasets

Virtual Archaeology Review, 8(17): 31-41, 2017 40 RECREATING A MEDIEVAL URBAN SCENE WITH VIRTUAL INTELLIGENT CHARACTERS: STEPS TO CREATE THE COMPLETE SCENARIO

Rodrigues, N., Magalhães, L.; Moura, J. P, & Chalmers, A. (2014). Reconstruction and generation of virtual heritage sites. Digital Applications in Archaeology and Cultural Heritage, 1(3-4), 92–102. http://dx.doi.org/10.1016/j.daach.2014.06.003 Rua, H., & Gil, A. (2014). Automation in heritage – parametric and associative design strategies to model inaccessible monuments: the case-study of Eighteenth-Century Lisbon Águas Livres Aqueduct. Digital Applications in Archaeology and Cultural Heritage, 1(3–4), 82–91. http://dx.doi.org/10.1016/j.daach.2014.06.002 Saldaña, M. (2015). An integrated approach to the procedural modeling of ancient cities and buildings. Digital Scholarship in the Humanities, 30(Suppl. 1), i148–i163. http://dx.doi.org/10.1016/j.daach.2014.06.002 SecondLife (2016). Retrieved from http://secondlife.com Trescak, T., Bogdanovych, A., & Simoff, S. (2014). City of Uruk 3000 B.C: Using genetic algorithms, dynamic planning and crowd simulation to re-enact everyday life of Ancient Sumerians. In Conference on the Simulation of the Past to Understand Human History (SPUH 2014). Unity (2016). Retrieved in January 2016 from http://unity3d.com

Virtual Archaeology Review, 8(17): 31-41, 2017 41

Virtual Archaeology Review, 8(17): 42-48, 2017 http://dx.doi.org/10.4995/var.2017.5845 © UPV, SEAV, 2015

Received: May 30, 2016 Accepted: June 6, 2017

DESIGN OF A VIRTUAL TOUR FOR THE ENHANCEMENT OF LLÍRIA’S ARCHITECTURAL AND URBAN HERITAGE AND ITS SURROUNDINGS DISEÑO DE UN TOUR VIRTUAL PARA LA PUESTA EN VALOR DEL PATRIMONIO ARQUITECTÓNICO Y URBANO DE LIRIA Y SU ENTORNO José Miguel Maícas, María José Viñals* Departamento Ingeniería Cartográfica, Geodesia y Fotogrametría, Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain. [email protected]; [email protected]

Abstract: The Information Technology and Communications (ICT) have revolutionized the way to present and promote the heritage sites. These ICT also offer scholars, students and visitors unprecedented access to architectural, historical, geographical, archaeological, iconographical and anthropological data, among other. It is noted also that virtual heritage environments are inherently fascinating and possess essential properties to have a positive effect on supporting heritage conservation and education. This paper is concerned with the potential of these ICT developments for improving the enhancement of the heritage sites of the town of Llíria and its surroundings (Valencia, Spain) by mean of a virtual tour (“Edeta 360º”) based on 360º panorama photos. The “Edeta 360º” virtual tour is an immersive application that places the viewers inside the image, enabling them to significantly enhance position awareness and providing the highest level of functionality for viewing, capturing and analysing virtual data. It can appropriately and effectively be utilised to facilitate intellectual and physical access to public by bringing knowledge, awareness and appreciation about the heritage of Llíria while, at the same time, authenticity is preserved. The undertaking method to create this interactive virtual tour is based on an easy procedure with off-the-shelf equipment and using both freely available software to address the process of photo stitching that combines multiple photographic images with overlapping fields of view to produce a segmented panorama or high-resolution image. Each panorama contains hotspots that enable the users to further explore the surroundings. The virtual tour provides the user the ability to navigate a scene through the rotation and zoom functions. This application results very appealing and has been adopted as a mean for information, dissemination, education and tourism purposes. Key words: ICT and cultural heritage, virtual tour, panoramic images

Resumen: Las Tecnologías de la Información y la Comunicación (TICs) han revolucionado la manera de presentar y promover los lugares patrimoniales. Las TICs ofrecen la posibilidad a académicos, estudiantes y visitantes de acceder, como nunca antes había sucedido, a información arquitectónica, histórica, geográfica, arqueológica, iconográfica y antropológica. Hay también que resaltar también la fascinación y los efectos positivos que los ambientes virtuales desempeñan en la conservación y la educación. Este artículo se centra en el potencial que tienen las TICs para mejorar la puesta en valor de los elementos patrimoniales de la ciudad de Liria y sus alrededores (Valencia, España) gracias al desarrollo de un tour virtual (“Edeta 360º) realizado a partir de fotografías panorámicas de 360º. El tour virtual “Edeta 360º” es una aplicación inmersiva que sitúa al observador dentro de la imagen, permitiéndole tomar conciencia de la situación de forma significativa y proveyéndole de un alto nivel de funcionalidad para poder ver, capturar y analizar los datos virtuales. Esta aplicación es apropiada y eficiente para facilitar el acceso intelectual y físico al público ya que fomenta el conocimiento, la concienciación y el aprecio por el patrimonio de Liria, a la vez que preserva la autenticidad del lugar. El método utilizado para llevar a cabo este tour virtual interactivo se ha basado en un procedimiento sencillo, utilizando equipamiento propio y software libre que combina múltiples imágenes fotográficas superponiendo campos de visión para producir una imagen de alta resolución. Cada panorama contiene puntos de interés que permiten al usuario explorar en detalle los alrededores. El tour virtual brinda al usuario la posibilidad de navegar en la escena a través de las funciones de rotación y zoom. Esta aplicación resulta muy atractiva y ha sido adoptada como medio de información, divulgación, educación y turismo. Palabras clave: TICs y patrimonio cultural, tour virtual, imágenes panorámicas

1. Introduction and objective 2011; Rogerio-Candelera, 2014), with special attention given to architectural heritage and urban spaces. In the last few decades, numerous studies have researched into how to apply Information and More recently, with the development of spatial simulation Communication Technologies (ICT) to cultural heritage and visualisation tools for spaces that cannot be (i.e. Monod & Klein, 2005; Styliaras, Koukopoulos & accessed easily, several applications have focused on Lazarinis, 2010; Hermon & Kalisperis, 2011; Ott & Pozzi, studying immersive and realistic forms of communication

*Corresponding author: María José Viñals, [email protected] 42 DESIGN OF A VIRTUAL TOUR FOR THE ENHANCEMENT OF LLÍRIA’S ARCHITECTURAL AND URBAN HERITAGE AND ITS SURROUNDINGS which provide public access to the contents of cultural 2. The town of Llíria and its surroundings heritage. According to Niccolucci (2002), virtual heritage productions may be ideal in responding to a need for a Llíria, the county town of the Camp de Turia, is located fashionable synergy between scientific enquiry, 25 km away from Valencia between the Mediterranean technology, art, and everyday life, and, consequently, coastal strip and the Iberian mountains. The municipal influence more serious cultural demand. district covers an area of 228 km2, being one of largest in the province. Among these applications, virtual environments have been widely used in the field of cultural heritage. They This strategic location has favoured the settlement of allow the general public to appreciate remote (in space several historical civilisations going back to pre-Roman and time) cultural assets with an immersive experience. times (at least the second millennium BC). This is the case of many virtual museum applications Archaeological remains of all of these civilisations (Styliania, Fotisb, Kostasa, & Petrosa, 2009; Miguélez (Iberians, Romans Visigoths, Byzantines, Muslims and Fernández, 2013; Reffat & Nofal, 2013), which offer the Christians) are scattered throughout the entire municipal opportunity to explore a remote site by manipulating district and many of them are also concentrated in the (rotating, enlarging, etc.) fragile precious objects with no historical town itself. risk of damaging them. It is also the case of Previous work was done in this issue (Maicas & Viñals, archaeological sites, where reconstructing damaged 2016). Currently an extended version of the application structures or remains allows users to gain a realistic is presented. Edeta 360º focuses on the process of overall vision of the ancient buildings (e.g. Rodrigues, virtualisation of the historical buildings located in the Magalhaes, Moura, & Chalmers, 2008; Ercek, Viviers, & historical town of Llíria, without excluding the possibility Warzée, 2010; Guidi, Russo, & Angheleddu, 2014). of extending the study area in further works. Among the different virtual approaches, virtual tours (VT) The decision to utilise ICTs was justified by the are one of the most popular productions (e.g. Bastanlar imbalance that existed between the large number of et al., 2008; Wessels, Ruther, Bhurtha, & Schröeder, heritage assets to be managed and the limited staffing 2014). resources. This makes it impossible to keep all the A virtual tour consists of a panoramic photography monuments open with a regular schedule for visits. application that allows users to use a mouse to interact Another factor, as mentioned earlier, is the geographical with the panorama by, for example, rotating it in all dispersal of the assets in a large district, which also directions, from the floor to the ceiling, to go from one forces the Town Council to keep many sites closed to room to another, to zoom on the image, etc. Additionally, the public. multimedia information and hyperlinks can be Additionally, accessibility is affected by difficulties related incorporated in order to provide more in-depth to the urban topographical features (slopes, grades, information so that the final presentation is enriched. etc.), and also because the private owners of some This kind of application is currently being used historical buildings do not allow visitors on their extensively around the world because it is able to properties. On the other hand, many old buildings also provide views of a wide variety of spatial data (realistic offer difficulties when it comes to accessing certain vision of building and surroundings) in a single rooms, such as the bell towers, upper floors, etc. environment, and to do so in a very attractive, interactive and meaningful way, thereby facilitating site cognition 3. Methodology and learning. There are different ways to create a 360º virtual tour. The objective of this paper is to describe the work This work employs the easiest procedure in order to use undertaken to create an interactive virtual tour based on off-the-shelf equipment and mostly freely available panoramic photos for the town of Llíria and its software, while still ensuring the quality of the final surroundings, called “Edeta 360º” (the name of the old product. The simulation environment is based on the Iberian town). same concept as that used by Google Street View; This work is a practical application developed within namely, it consists of a set of spherical images that the framework of the International Postgraduate MSc capture the whole environment around the data Programme "Building Heritage Conservation” at the collection point where they are performed. Universitat Politècnica de València. It represents a The selection of the different heritage assets and/or case study of the use of technology that aims to environments was the first step in the elaboration of the provide the general public with access to Llíria’s virtual tour. The criteria applied to select these heritage heritage, thereby enabling them to learn about it and elements were primarily: significance, singularity, appreciate it. At the same time, it serves as a mean to representativeness and uniqueness, but also physical disseminate and share the wide and rich heritage of accessibility. this municipality, and to engage users in a real experience of the town of Llíria. Thus, eighteen visitable cultural assets (Fig. 1) were included in the virtual tour. These cultural assets are It has, at all times, followed the London Charter for the already recognised as “Assets of Cultural Interest” Computer-Based Visualisation of Cultural Heritage (Bienes de Interés Cultural – BIC) by the Spanish (2009), which establishes internationally-recognised national heritage protection law and they are also principles for the use of computer-based visualisation by “Assets of Local Relevance” (Bienes de Relevancia researchers, educators and cultural heritage Local –BRL) in accordance with regional law. All of them organisations1. are protected by the Llirian Urban Planning Programme (Catálogo de Bienes y Espacios Protegidos del Plan General de Ordenación Urbana – PGOU). 1 http://www.londoncharter.org/ (Accessed January 13th, 2016)

Virtual Archaeology Review, 8(17): 42-48, 2017 43

MAÍCAS & VIÑALS, 2017

Vila Vella of circumference, it begins another turn of 30º from the  Ayuntamiento de Llíria, Casa Consistorial (Ca la Vila) main direction, and so on successively until the  Baños Árabes circumference is completed.  Horno de Pancocer (Forn de la vila)  Iglesia del Buen Pastor (Hospital medieval)  Iglesia de la Sangre  Parroquia de la Asunción de Nuestra Señora  Iglesia de Nuestra Señora del Remedio  Iglesia de la Mare de Déu  Muralla Medieval de Llíria Tossal de Sant Miquel  Poblado Ibérico Cerro San Miguel, Edeta  Real Monasterio de San Miguel Edeta romana  Colegio San Vicente Ferrer  Mausoleos Romanos El “Raval”  Almudín  Ermita de Santa Bárbara  Parroquia de San Francisco de Asís El “Pla de l’Arc” Figure 2: Spherical coordinates (radius and two angles).  Ermita de San Vicente (Santuario de San Vicente)  Parroquia de María Madre de la Iglesia A total of 6,000 photos of Llíria were captured during 12 field campaigns between January and June 2015. Figure 1: List of the Llirian cultural assets included in the virtual tour. The result of this procedure is a spherical photo- mosaic, which consists of the retroprojection of the Secondly, all information related to these assets was photographs on a spherical surface. In our case, 70 recorded from documentary and bibliographic sources panoramas of Llíria were created, which is a number (digital, graphical and written materials), so that they that is sufficient to create a fluid route. Additionally, 20 could be enhanced virtually. This material includes aerial aerial images from Google Earth and Iberprix were images of the town, old pictures, historical texts, artistic imported to create site maps. drawings, etc. 3.2. Rendering the spherical images onto flat Another task was to carry out walks around the territory and the town in order to become familiar on a first-hand 2D surfaces basis with the reality of the selected heritage assets. First of all, some photo edits or corrections were The steps followed in the process of developing the performed; then, luminosity and colour-level adjustments virtual tour are described below. were made in order to obtain homogeneous images, and optical lens distortions were also corrected. 3.1. Taking photos The most commonly available rendering options for virtual tours, according to Ippoliti, Calvano & Mores The purpose of this task is to create the panorama view (2014), are spherical geometric projections (also called by applying 360º photography techniques in order to equirectangular projection). This is really a capture the entire surroundings of a location. cartographic representation that consists in transposing To this end, at each cultural asset a central shooting points of the sphere onto a straight cylindrical surface position was established to guarantee the same focal imagined to be tangent to the sphere at its greatest length and taking into account the radius on which it will parallel. swing so as to ensure there was enough image overlap (at least 25-30%) to allow the photos to be stitched together, as suggested by Fangi (2006). The spherical coordinates are established with a radial distance (r) and two angles (, ), as depicted in Figure 2. The radius has to be as small as possible; the first angle (), with an opening of 0º to 360º, rotates in a horizontal plane. The third coordinate is another angle () with a value ranging from -90° to 90° in the vertical plane. The number of shots in this case study was very high because fisheye lens are not used in this procedure. The images did not contain any moving objects or people because these could cause errors in the scene. Photographs were taken based on the coordinates 0.1, 0º, -90º; namely, a small radius pointing in the main direction and at the floor. From here, the third coordinate at 30° was increased to complete the 180º. After this arc Figure 3: Texture overlapped onto a sphere.

Virtual Archaeology Review, 8(17): 42-48, 2017 44

DESIGN OF A VIRTUAL TOUR FOR THE ENHANCEMENT OF LLÍRIA’S ARCHITECTURAL AND URBAN HERITAGE AND ITS SURROUNDINGS

Figure 4: La Mare de Déu texture 360º.

3.3. Stitching photos to create 360º panorama and interior. In order to contemplate the building from images different angles, the spheres needed to be able to walk around were constructed. Afterwards, the interior of the This was made performed by using the open source building can be visited. The spheres are connected panorama stitcher HUGIN 2014.0.0, developed by Pablo consistently, and the inside of the building can only be d'Angelo. This is a freely available software application accessed through a sphere that includes a door. that is licensed under the GPL, and runs on Microsoft Windows® and GNU/Linux®. This program imports the photos into a raster-based application, superimposes them two by two, and generates a single panoramic image of the deployed sphere that can be mapped onto planar surfaces. In these images, like a flat map of the Earth, the most remote areas are highly distorted; the image will only be seen properly when the texture is overlapped on a sphere as seen in Figure 3. The final result of this procedure was a collection of high quality 5000 x 2500 pixel images (Fig. 4). This is an acceptable resolution that allows distant elements to be zoomed in on without loss of quality.

3.4. Converting a 360º panorama into a virtual Figure 5: Locations of spheres. tour The panorama generated by the stitching software was 3.5. Adding extras to the virtual visit imported into the Easypano Tourweaver 7.90® Navigation instruction texts, heritage information software application to produce the visitable and documents (different plans showing the evolution of the interactive urban virtual tour. It allows the panorama to building over time, documentary texts, etc.), audio files, be set in motion and the creation of a navigation images (old pictures, etc.) are included in the application interface that allows users to walk through the scene as pop-up windows. This information enriches the virtual virtually. Each spherical scene is connected to another tour, allowing visitors to gain more comprehensive by a link. With this spatial configuration, the user does knowledge about each heritage asset. not move continuously but discretely, jumping from one fixed position to the next, and having viewsheds in any 4. Results: Appearance of Edeta 360º direction, but from a finite number of locations (Fig. 5) because the user is confined to specific viewpoints. Edeta 360º opens when you right-click the virtual tour; it Nevertheless, this interactive application allows the then shows a contextual menu (Fig. 6). user to take control over the panoramic visit by arranging different kinds of movements (rotation, The second screen displayed consists of an introductory inclination, close-ups, distancing, etc.). text (Fig. 7) with instructions on how to navigate and which also invites the user to enter. To further explore the site, the visitor can move on to investigate other structures and the landscape (from one After clicking on the access button, navigation starts with scene to another) by way of clickable invisible polygonal an urban aerial photo (Fig. 8) of the historical town of hot spots (opacity 0%), which will be triggered when a Llíria and related zooming images of the different pointing device (usually a mouse) is moved over another virtually enhanced heritage sites and surroundings: La heritage element whether visual contact exists between Vila Vella (Fig. 9), Edeta Romana, Tossal de Sant them. The images to be connected can be both exterior Miquel, El Raval and Pla de l’Arc. The visitor can select

Virtual Archaeology Review, 8(17): 42-48, 2017 45

MAÍCAS & VIÑALS, 2017 one of them by clicking on the area and, afterwards, a historical building can be chosen by clicking on it; then, a first image of the main access door of the building appears.

Figure 9: Detail of historical centre La Vila Vella.

When the cursor slides over streets, doors, stairs, etc., and becomes a hand, another connected panoramic sphere can be opened if clicked on. The cursor can also be transformed without any navigation motion. This occurs if there is additional pop-up information associated with the object. In Figure 10, the Iglesia de la Sangre is presented as an example (Fig. 10a), and showing the appearance of the Figure 6: Edeta 360º loading screen: Contextual menu. additional pop-up panel for the Bell Tower (Fig. 10b).

(a)

(b) Figure 7: Edeta 360º introductory screen. Figure 10: Iglesia de la Sangre with Bell Tower: a) Presentation screen of the church; b) Additional pop-up window.

The building can be seen in all directions; it can be walked around or the inside can be accessed (Fig. 11). Indoor, it can also be visited virtually and in-depth information about it can be obtained by deploying the pop-up windows.

5. Discussion and concluding remarks As final thoughts about the project that was developed, the following comments can be highlighted. Firstly, it must be mentioned once again that virtual heritage environments need to possess certain essential properties to have a positive effect on the general public.

Beyond the well-established methods that have Figure 8: Aerial photo of the centre of Llíria. traditionally been employed in the field of cultural

Virtual Archaeology Review, 8(17): 42-48, 2017 46

DESIGN OF A VIRTUAL TOUR FOR THE ENHANCEMENT OF LLÍRIA’S ARCHITECTURAL AND URBAN HERITAGE AND ITS SURROUNDINGS

Additionally, another contribution of Edeta 360º is the overall vision, featuring the heritage of a destination from an urbanistic perspective rather than a view of single elements. This presentation makes it possible for the user to know all of the assets that make up the whole site. This vision is necessary to understand the history of the place and this knowledge would be temporally and spatially impossible to obtain in many tourist destinations unless offered through a virtual tour. This panoramic tour is available online, at www.edeta360.com. Regarding the maintenance and updating of the application, it should be noted that a

project based on ICTs, because of their rapid evolution, (a) risks becoming obsolete in a short time. Therefore, it is necessary to update it regularly, not only in terms of the content but also as regards the devices from which you can access it. In this case, updates are performed by ftp, with the FileZilla program. In the last update, an app available on Play Store (Edeta 360º) was created that allows access to this content from an icon on a smartphone or tablet without having to type the URL in a browser. Another update that is being implemented at the moment is the inclusion of 360º videos, thus allowing the user to have more immersive views. These videos are also compatible with virtual reality cameras. Furthermore, to reach a larger number of users, this application would need to be translated into other (b) languages. Figure 11: Iglesia de la Sangre: (a) Screen with navigation On the other hand, it is noteworthy that the current options; (b) Pop-up complementary windows. approach with simple off-the-shelf equipment is an important strength of the project. It thus gives an idea of heritage to date, ICTs, and particularly virtual tours, the broad potential of its extrapolation to other similar have become increasingly more popular tools to heritage assets. Moreover, it must be remembered that achieve heritage enhancement, tourism experience this process is open to possible technological upgrades development and dissemination. This is because virtual and expansions in the study area, so everything carried tours have the ability to arouse fascination far beyond a out so far can be considered a good starting point. tourist brochure or any printed material. For this However, it must be recognised that it is necessary to reason, it is a widely used application in cultural events look deeper into how to strengthen the emotional and tourism fairs. dimension of these presentations. Some authors, like On the other hand, an online virtual tour can effectively Roussou (2008), have already opened up this debate. contribute to the recreational experience in the sense For this reason, future updates will include people, in that Viñals, Morant & Teruel (2014) noted. Hence, order to better engage and empathise with the user. besides the onsite phase of visitation, the virtual tour With all these improvements, Edeta 360º will be far more has a high potential to generate motivation and popular and appealing. expectations in the phase prior to the visit. A virtual Finally, it must be noted that local heritage online tour can also consolidate feelings of administrations are in the best position to lead the appreciation after the visit. design of the virtual reality productions that are needed However, it should be noted that Llíria can be visited to conserve, enhance and make their heritage more virtually by Google Street View, but this application dynamic because local communities are the first to does not include many of the areas shown in Edeta benefit from these actions. 360º or the interior of buildings. Other specific advantages have been identified, such as its Acknowledgements contribution to address accessibility issues (private The authors of this work wish to thank Llíria Town ownership of the heritage, physical difficulties to visit Council and other institutions for supporting the some places, etc.). This application also includes development of Edeta 360º. We are also grateful for additional information such as text, audio, evolution of the scientific peer reviewing work carried out by the buildings over time, among others. Dr. Cristina Portalés.

References Bastanlar, Y., Grammalidis, N., Zabulis, X., Yilmaz, E., Yardimci, Y., & Triantafyllidis, G. (2008). 3D reconstruction for a cultural heritage virtual tour system. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 37-B5, 1023–1036. Ercek, R., Viviers, D. & Warzée, N. (2010). 3D reconstruction and digitalization of an archaeological site, Itanos, Crete. Virtual Archaeology Review, 1(1), 81–85. http://dx.doi.org/10.4995/var.2010.4794

Virtual Archaeology Review, 8(17): 42-48, 2017 47

MAÍCAS & VIÑALS, 2017

Fangi, G. (2006). Investigation on the suitability of the spherical panoramas by Realviz Stitcher for metric purposes. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 36-5, 25–27. Guidi, G., Russo, M., & Angheleddu, D. (2014). 3D survey and virtual reconstruction of archeological sites. Digital Applications in Archaeology and Cultural Heritage, 1(2), 55–69. http://dx.doi.org/10.1016/j.daach.2014.01.001 Hermon, S. & Kalisperis, L. (2011). Between the real and the virtual: 3D visualization in the cultural heritage domain - expectations and prospects. Virtual Archaeology Review, 2(4), 59–63. http://dx.doi.org/10.4995/var.2011.4556 Ippoliti, E., Calvano, M. & Mores, L. (2014). 2.5/3D models for the enhancement of architectural-urban heritage. A virtual tour of design of the Facist Headquarters in Littoria. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2-5, 189–196. http://dx.doi.org/10.5194/isprsannals-II-5-189-2014 Maicas, J. M., & Viñals, M. J. (2016). Edeta 360º virtual tour for visiting the heritage of Lliria (Spain). Lerma, J.L. & Cabrelles, M. (eds.): Proceedings of the Archaeologica 2.0. 8th International Congress on Archaeology, Computer Graphics, Cultural Heritage and Innovation. Valencia, 5-7 September, 2016. Ed. Universitat Politècnica de València, 376–378.

Miguélez Fernández, L. (2013). Tour virtual por la red de museos de Gijón. Tesina final de Master. Universidad de Oviedo. Monod, E., & Klein, H. K. (2005). A phenomenological evaluation framework for cultural heritage interpretation: from e- HS to Heideggers’ historicity. Proceedings of the eleventh Americas Conference on Information Systems, Omaha, USA, 2870–2877. Niccolucci, F. (Ed.) (2002). Virtual archaeology: Proceedings of the VAST 2000 Euroconference, Arezzo 24-25 November 2000. Oxford: Archaeopress. Ott, M., & Pozzi, F. (2011). Towards a new era for cultural heritage education: Discussing the role of ICT. Computers in Human Behavior, 27(4), 1365–1371. http://dx.doi.org/10.1016/j.chb.2010.07.031 Reffat, R. M., & Nofal, E. M. (2013). Effective communication with cultural heritage using virtual technologies. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40-5W2, 519–524. http://dx.doi.org/10.5194/isprsarchives-XL-5-W2-519-2013

Rodrigues, N., Magalhaes, L. G., Moura, J. P., & Chalmers, A. (2008). Automatic reconstruction of virtual heritage sites. Proceedings International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage. Ed. The Eurographics Association. http://dx.doi.org/10.2312/VAST.VAST08.039-046 Rogerio-Candelera, M. A. (Ed.) (2014). Science, Technology and Cultural Heritage. London: CRC Press/Balkema. Roussou, M. (2008). The components of engagement in virtual heritage environments. Proceedings of new heritage: beyond verisimilitude. Conference on Cultural heritage and New Media, Hong Kong, 265–283. Styliania, S., Fotisb, L., Kostasa, K., & Petrosa, P. (2009). Virtual museums, a survey and some issues for consideration. Journal of Cultural Heritage, 10(4), 520–528. Styliaras, G., Koukopoulos, D., & Lazarinis, F. (Eds.) (2010). Handbook of research on technologies and cultural heritage: Applications and environments. (pp. 1-592). Hershey, PA: IGI Global. http://dx.doi.org/10.4018/978-1- 60960-044-0 Viñals, M. J., Morant, M., & Teruel, L. (2014). Confort psicológico y experiencia turística. Casos de estudio de espacios naturales protegidos de la Comunidad Valenciana (España). Boletín de la AGE, 65, 293–316.

Wessels, S., Ruther, H., Bhurtha, R., & Schröeder, R. (2014). Design and creation of a 3D virtual tour of the world heritage site of Petra, Jordan. Proceedings of AfricaGeo (pp. 1–12).

Virtual Archaeology Review, 8(17): 42-48, 2017 48

Virtual Archaeology Review, 8(17): 49-55, 2017 http://dx.doi.org/10.4995/var.2016.4650 © UPV, SEAV, 2015

Received: February 5, 2016 Accepted: July 19, 2016

DIGITIZATION OF RELIGIOUS ARTIFACTS WITH A STRUCTURED LIGHT SCANNER ESCANEADO DE ARTEFACTOS RELIGIOSOS CON UN ESCÁNER DE LUZ ESTRUCTURADA Alejandro Graciano*, Lidia Ortega, Rafael J. Segura, Francisco R. Feito Department of Computer Science, University of Jaén, Spain. [email protected]; [email protected]; [email protected]; [email protected]

Abstract: The digitization process for religious artifacts is subject to inherent difficulties often ignored in theoretical models or pipelines. In this paper we aim to describe these problems, which are present in practical environments such as temples and churches, using white light scanners instead of other common devices or technologies such as laser scanners and photogrammetry. Our case study is based on the digitization of two religious statues belonging to a Catholic brotherhood located in a village of the Province of Jaén (Spain), one of them presenting especially several limitations. After performing the scanning process with a portable hand-held scanner, the images captured were processed until the final models were acquired. On the basis of the results obtained, we discuss the problems arising after using well-known procedures for the reconstruction of 3D models, their causes and some possible solutions to achieving a correct digitization. It should be noted that it is not the aim of this study to establish procedures for the digitization of religious artifacts, but rather to transmit the inherent constraints of these types of scenes. Key words: 3D scanning, white light scanner, religious artifacts, cultural heritage, digitization

Resumen: Los modelos teóricos de digitalización 3D no tienen en cuenta problemas de escenarios específicos como el de la digitalización de objetos religiosos. El objetivo de este artículo es describir estos problemas presentes en entornos prácticos como templos o iglesias usando un escáner de luz blanca en lugar de otros instrumentos o tecnologías comunes como los escáneres láser y la fotogrametría. Nuestro caso de estudio se basa en la digitalización de dos tallas religiosas pertenecientes a una cofradía de una localidad de la provincia de Jaén (España), siendo especialmente restrictiva una de ellas. Tras el procedimiento de escaneado llevado a cabo con un escáner de mano, las capturas se procesaron usando procedimientos clásicos de reconstrucción de modelos 3D hasta obtener los resultados finales. Basándonos en los resultados obtenidos realizamos una discusión de los problemas, causas y posibles soluciones para llevar a cabo una correcta digitalización. Cabe destacar que el objetivo del artículo más que establecer un flujo de trabajo es el de presentar las restricciones que presentan este tipo de entornos. Palabras clave: escaneado 3D, escáner de luz blanca, artefactos religiosos, patrimonio cultural, digitalización

1. Introduction temples or artifacts in an accurate way. For instance, there are recent papers focused on this technique such The digitization and reconstruction of cultural and as the one presented by Rodriguez-Gonzálvez, religious heritage are providing artists, historians and the Nocerino, Menna, Minto, & Remondino (2013) and the general public with a new methodology for learning work of Menna et al. (2016). However, this technique is about and researching these items. These novel less suitable for the digitization of small pieces, such as technologies allow the dissemination of cultural heritage in our case study (Nicolae, Nocerino, Menna, & to the general public through virtual museums and tours Remondino, 2014). (Styliani, Fotis, Kostas, & Petros, 2009; Kiourt et al., 2015); the restoration and conservation of heritage In this paper we include the experiences of scanning (Lanitis, Stylianou, & Voutounos, 2012) and research religious artifacts in a real-world environment and the into other cultures or eras (Hermon et al. 2013). problems involved. The rest of the article is structured as follows: in Section 2 we present an overview of the There are several types of scanners that can be used to state-of-the-art related to structured light scanning digitize the cultural heritage such as white light religious artifacts. Section 3 depicts the hardware and scanners, laser scanners and computerized software used during the scanning process. In tomographies (CT). Many of these former scanners are Section 4, we detail the process carried out and the portable or hand-held devices, which makes the data problems encountered. Then, we discuss the results acquisition much easier. Photogrammetric methods are obtained in Section 5. Finally, we conclude the paper in also widely used in order to digitize and document Section 6.

* Corresponding author: Alejandro Graciano, [email protected] 49 GRACIANO et al., 2017

2. Previous work Artec Studio 9 software was utilized for the processing of the scans. The features provided by this software are Some religious heritage scanning projects have been sufficient for a common scanning process. These performed, mainly inside temples or churches. Some features include sharp fusion and mesh simplification, examples are those carried out in the Church of San among others (Cignoni, Montani, & Scopigno, 1998). Francisco, the Church of A Coruña (Spain) (Pérez & Unfortunately this tool is a commercial software solution, Robleda, 2015), the Cathedral of Notre-Dame des and therefore it is not an easy task to know the concrete Amiens (Crombez, Caron, & Mouaddib, 2015) and the algorithms used. Recently Artec Group released a new Cathedral of Jaén (Soria, Ortega, Feito & Barroso, version of this software, Artec Studio 10. This version 2015). Some of these projects related to religious contains a faster alignment algorithm and an artifacts are performed outside of temples, as in the case improvement in the geometry and texture editing of the ancient city of Herculaneum where the amazon process. Figure 2 shows a screenshot of the software woman’s head was scanned (Happa et al., 2009). In package. 2003 some projects were carried out for the digitization and subsequent restoration of several sculptures located In order to test the quality of the scanner we carried out in the Cathedral of Santa Mara in Florence and the a previous test with two sample figures (a tooth model Basilica of San Pietro in Vincoli, amongst many other sculpted in clay and a clown model). As shown in Italian temples. For example, in this project they digitized Figures 3 and 4, the perceived visual quality of the the statue of the Prophet Hababuc sculpted by Donatello results is reasonably accurate. This software was and the sculpture built by Nanni di Banco, The Four executed in a computer with an Intel Core i7-4600U Crowned Saints. These statues are located in processor and 16 GB RAM. Orsanmichele (Florence) (Salimbeni, Pini, & Siano, 2003). In (Van Gasteren, 2013) a project is presented for the reproduction of two sculptures, Saint Teresa de Jesús and Christ Tied to the Column, both located in the Convent of Santa Teresa (Ávila, Spain). More recently Díaz-Marín et al. (2015) digitized the fragments of an archaeological Terracotta statue in order to obtain a 3D model of the whole statue. Other digitization processes of cultural artifacts can be performed by means of techniques such as Structure from Motion (SfM) (Barsanti, Micoli & Guidi, 2013; Nabil

& Saleh, 2014) and photogrammetry (Remondino, 2011; Rinaudo, Chiabrando, Lingua & Spanò, 2012; Dall’ Asta, Figure 2: Screenshot of Artec Studio software. Bruno, Bigliardi, Zerbi & Roncella, 2016).

3. Material

Artec (Artec Group, 2016) markets two hand-held scanners. On the one hand the Eva version, that is suitable for medium size models (as in our case study), and on the other hand the Spider version which is more appropriate for working with small models. The 3D scanner used in this study is the Artec Eva (Fig. 1). This hand-held structured-light scanner captures 3D images with a frequency of 16 frames per second. Artec Eva is not completely appropriate for scanning small figures or details even though the scanner has a huge resolution, 0.5 mm. The minimal distance needed to scan is between 0.4 m and 1 m, subject to the illumination in the particular scene. The capture resolution is 1.3 MP. (a) (b) Figure 3: Clown model: a) 3D scanned model; b) real model.

(a) (b) Figure 1: Hand-held scanner Artec Eva. Figure 4: Tooth model: a) 3D scanned model; b) figure modeled in clay.

Virtual Archaeology Review, 8(17): 49-55, 2017 50 DIGITIZATION OF RELIGIOUS ARTIFACTS WITH A STRUCTURED LIGHT SCANNER

4. Digitization process 4.1. Problems involved in scanning religious artifacts The digitization process of 3D models from the scans must follow a succession of steps (pipeline) in order to In practice we may find ourselves with some problems achieve proper results. The following pipeline was when scanning religious artifacts. Many of these introduced by Bernardini & Rushmeier (2002). This problems, such as poor illumination or elements with pipeline consists of two process streams; one for the little geometry (for instance, thin or small objects), can geometry of the model and another for the appearance be found in many other scenarios, but the constraints properties of the model surface (Fig. 5). that this specific scene provides can increase the The process is totally sequential, as indicated by the difficulty in achieving proper scans. For instance, small dotted line. It is very usual to obtain a feedback between features can be removed from the statue and scanned both streams in order to improve the quality and separately; but this often cannot be allowed due to efficiency of the processing of each type of data. This religious reasons, as explained below. Although some of model assumes perfect conditions for data acquisition the problems can be solved by means of data but, in fact, it may often be deficient. These difficulties postprocessing tasks, poor data capture conditions can may be due to the device used (e.g. scanner accuracy), require the repetition of the data acquisition procedure. the model to be digitized (shining/reflective material) and In our case, in a first test run, we noted the following the environmental conditions (e.g. adequate illumination difficulties during the capture process with a structured or freedom of movement). light 3D scanner.  Poor illumination. Generally, these sculptures are located either in churches or in cathedrals, places where the illumination of the ambient is usually insufficient. The inclusion of any other type of illumination such as focal point lights or directional lights is not suitable due to the generation of shadows in the model. Shadows can be a problem because of two reasons: on the one hand, the geometry of the model may not be captured completely since many white laser scanners do not recognize parts of the model that are overly dark. On the other hand, the output textures can contain different shades depending on these shadows. The final result can be improved to a certain extent by performing a 3D reconstruction of the illumination in order to edit the light sources (López-Moreno, Hadap, Reinhard, & Gutierrez, 2010).  Shiny ornaments. Some vestments and crowns, or sceptres, usually contain metallic or golden details. Scanning these kinds of materials can often leave gaps in the model, mainly because of the reflections produced by these kinds of objects. As stated earlier, also relevant are the cases in which the addition of some types of scanners, such as laser scanners, may not recognize dark or black parts of the model. This problem can be partially solved by Figure 5: Pipeline for the digitization of an object in order to adjusting the light sensitivity for capturing highly achieve a 3D model. illuminated parts; however, some darker parts may not finally be scanned.

(a) (b) (c) (d) (e) Figure 6: Example in digitization pipeline: a) raw scans; b) scans registration; c) mesh fusion; d) mesh simplification; e) textured model.

Virtual Archaeology Review, 8(17): 49-55, 2017 51 GRACIANO et al., 2017

 Difficult access. The movement of these statues process of the scans captured with a different outside the church or cathedral is often complicated orientation might be necessary (e.g. if the model is because of administrative or religious issues. rotated on an arbitrary axis). The Align step (so Furthermore, not everybody can manipulate this called in the Artec Studio) serves as a hint for the kind of artwork; it needs specialized personnel in registration. Pairs of conjugate points have to be order to prevent its damage. In addition to the selected manually in order to rotate and translate inability to move the statue to a room suitable for the the scan toward a reference scan. This step is scanning process, it is even possible that these optional and, in our case, was not required, as we sculptures cannot be removed from their altar. stated above, because we did not move the statue. These specific situations can make the access to 3. Removing unnecessary parts. It is necessary to several parts of the statue difficult or even remove superfluous information from the scans such impossible. as backgrounds or holders. This process is mainly  Elements with little geometry. Thin parts of the manual since the use of automatic algorithms may sculpture with little geometry may not be captured, not be accurate and may remove valid parts of the such as banner poles and fringes. This factor mainly model. Neglecting unnecessary geometry can depends on the minimal resolution of the device. provide less accumulated error in the following steps Commonly, these elements are scanned separately since there is less geometry to be processed. in order to be added to the final model later. 4. Removing outliers. Commonly, we may obtain  Direct manipulation of the sculpture. Due to the noise provided by glitters, an improper illumination delicateness and the religious meaning of this sort of or accumulated errors of previous steps. In figure, it is necessary that the handling be particular, a bad alignment which can produce undertaken by specialized staff, not only with regard overlapping areas between distinct scans. This to its movement but also to any kind of manipulation, process is performed again manually or like undressing it or removing any element such as semiautomatically, since it needs the supervision of crowns and banners. an expert in order to prevent deletion of valid parts of the model. 4.2. Digitization process of religious artifacts 5. Definition of the shape. Once the scans have been processed, it is necessary to create a single triangle In our practical case we digitized a wooden statue of mesh from the point cloud provided by the scans. Risen Christ placed in the Church of St. Mary in After this step it needs to be determinated that the Torredonjimeno (Jaén, Spain). These digitization mesh is completely watertight. services were requested by the Catholic brotherhood in order to disseminate their religious heritage. Due to the 6. Small object filtering. After these steps, it is very difficulty of the manipulation and the inability to scan the usual that the model has generated new large noise. whole statue, a replica of a statue of Holy Week was Classical algorithms for the detection of outliers may also digitized in order to offer a full preview of the not work properly in this case. Isolated elements scanning process. The replica was also provided by the whose size does not exceed a certain threshold Catholic brotherhood. The statue of Risen Christ is have been removed. Once again, it is necessary to approximately 1.60 m tall, whereas the replica is ensure that necessary parts have not been approximately 0.5 m tall. While the statue of Risen Christ removed. could not be moved from its altar, the replica could be located freely. In Figure 7 can be seen a moment during the scanning process. The digitization process for both statues was carried out following the pipeline described in Section 4. It is remarkable that the alignment step (included as a first step in the registration one) has been alleviated by the scanner used. The steps carried out are listed below: 1. Scanning. One of the golden rules of the scanning steps is to take the minimum number of scans that cover the whole target. In the case of the Risen Christ four scans were needed whilst only three scans were required for the replica. To ensure a good quality of the results, this step must be carried out carefully. Scanning the Risen Christ model took around one hour whilst the replica took only 30 minutes. 2. Registration of the scans. In this process all the overlapping captured scans must be fully referenced under the same common coordinate system in order to obtain a single scan (Besl & McKay, 1992; Digne, Morel, Audfray, & Lartigue, 2010). This process is the most expensive in terms of computational time of the overall pipeline, taking 4 h for the Risen Christ and 6 h for the replica. A previous alignment Figure 7: A photograph taken during the scanning process.

Virtual Archaeology Review, 8(17): 49-55, 2017 52 DIGITIZATION OF RELIGIOUS ARTIFACTS WITH A STRUCTURED LIGHT SCANNER

7. Postprocessing. Once we have a single model, there are additional tasks which need to be performed. For example, it is important to carry out a smoothing process to edges and corners, to remove flaws in the mesh such as small holes, and to simplify the number of triangles. The reason why we scanned at high resolution and then performed a decimation of triangles was because the triangle mesh decimation algorithms only simplify zones with small geometric changes, keeping the model’s fidelity. This ensures that zones with many details will be simplified little or not at all. In Table 1 we can see the number of initial and final triangles, as well as the file size for both models.

8. Texture mapping. It is necessary to perform a mapping between the image pixels (called texels) (a) (b) and the mesh triangles in order to provide the model with texture (Sander, Snyder, Gortler, & Hoppe, Figure 8: Final 3D models: a) Risen Christ; b) replica. 2001). Finally, an adjustment of parameters such as shining, contrast or saturation of the final texture is with difficult access due to the freedom of movement carried out. that it offers in comparison with a static setup. For instance, the digitization of parts close to a wall might An overview of the pipeline process can be seen in be more complicated since the camera setup might not Figure 6. fit sufficiently between the statue and the wall. Likewise, the inherent problem derived by a poor Table 1: Triangle mesh simplification illumination also affects the digitization performed by means of photogrammetric techniques. Also, the time Initial Final Initial size Final size Model triangles triangles (MB) (MB) required for taking the photographs can be significantly higher than the time for taking all the scans if a Risen Christ 1002289 204928 23 5 resolution of the 3D model is required comparable to Replica 803521 148497 18 4 that offered by a 3D scanner. However, the hardware for a photogrammetry study is much cheaper than for a 5. Discussion hand-held scanner device. The scanning of inaccessible parts could be achieved The final models of this process can be seen in Figure 8. using mirrors. This solution imposes a couple of Figures 8a and 8b show the resulting digitization of the restrictions: (1) the use of a laser scanner since white Risen Christ and the replica respectively. These results light scanners do not work properly with mirrors, and are obtained from a first contact with the sculptures and (2) the need to place the scanner in a fixed position in the indoor scene. The impossibility of removing one of order to reference the reflected scans with the other the statues from its altar increases the difficulty of scans. Due to the material used (a hand-held scanning the complete statue, for example, the back of white light scanner); this method could not be the statue. Neither could lateral parts be scanned since it employed. is required to keep a minimum distance between the model and the scanner. Furthermore, the spot and point lights located in the church did not help in capturing the scans. Additionally, the final textures showed small glares, as can be seen at the bottom part of the Risen Christ model (Fig. 8a). Finally, we decided to take scans only from the statue’s bust in order to test the process and show the results to the Catholic brotherhood. In addition, a complete process for the improvement of the replica captures was performed. Scanning with a structured (white) light scanner instead of with a laser scanner presents some advantages based on previous experiences scanning artifacts. In prior work we used a Polhemus FastSCAN I (Polhemus, 2016). This laser scanner uses an electromagnetic tracking technology to reference each scan into the same coordinate system. One of the main conclusions reached in that study was that this type of technology is not suitable for scanning sculptures of large size. This is due to the fact that the receptor can miss the signal from the electromagnetic transmitter if they are distant, which is not the case using a white light scanner. Figure 9: 3D printed model of the replica. A few elements, such Moreover, the benefits of scanning with this technique as the banner or parts of the crown, were neglected during the versus photogrammetry are clear in this scenario. A scanning process because of the printer resolution. The figure hand-held scanner is more appropriate to scanning parts has a height of 70 mm.

Virtual Archaeology Review, 8(17): 49-55, 2017 53 GRACIANO et al., 2017

Regarding texture correction, there are several problems presented during the scanning process were techniques, like “inpainting” (Criminisi, Pérez, & Toyama, partially solved. The software package used fixed 2004), which may repair the glares caused in textures by some issues such as the removal of outliers, the external light sources. These kinds of techniques are removal of unnecessary parts, the creation of a closed very interesting in order to manage the model in a 3D polygonal mesh, including its smoothing and modeling tool and to be able to calculate the correct simplification or texture correction. However, some shadows properly. problems regarding the physical location could not be solved. Therefore certain parts of one of the sculptures Finally, one of the next logical steps after the digitization of could not be scanned. In addition, several illumination the religious heritage is its impression in a 3D printer and issues dealing with glares and shadows remain still its commercialization (Fig. 9). It could be interesting to unsolved. introduce some watermarks to these figures in order to avoid their illicit exploitation. Some of these techniques In the near future, we expect to digitize the whole artifact are invasive and the figures reproduced could be partially in order to provide the Catholic brotherhood with a deformed, which would not be accepted by the devoted faithful representation of their sculptures and thus deal community. For this reason, non-visible or local with the unsolved issues. In this new digitization watermarking techniques could be applied (Wang, procedure we hope to be able to: (1) change the location Lavoué, Denis, & Baskurt, 2011; Luo & Bors, 2011). of the sculpture; and (2) scan it in a controlled environment, with proper illumination and without space 6. Conclusions and future work constraints. In this paper, we describe the practical process of scanning religious artifacts. This study does not intend to Acknowledgements create a workflow, but rather transmit the main problems This study has been partially supported by the encountered in possible working environments, as well Ministerio de Ciencia e Innovación and the European as some possible solutions. Union (via ERDF funds) under the research project This study has served as a first contact with scenes TIN2014-58218-R, and by the University of Jaén existing in either temples or churches. Some of the through the research project UJA2015/08/10.

References

Artec Group (2016). Artec 3D Scanners. Retrieved October 10, 2016, from http://www.artec3d.com/ Barsanti, S. G., Micoli, L. L., & Guidi, G. (2013). Quick textured mesh generation for massive 3D digitization of museum artifacts. In 1st Digital Heritage International Congress (DigitalHeritage), (pp. 197–200). Marseille, France. http://dx.doi.org/10.1109/DigitalHeritage.2013.6743732 Besl, P. J., & McKay, N. D. (1992). A method for registration of 3-D shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 14(2), 239–256. http://doi.org/10.1109/34.121791 Bernardini, F., & Rushmeier, H. (2002). The 3D model acquisition pipeline. Computer Graphics Forum, 21(2), 149–172. http://doi.org/10.1111/1467-8659.00574 Cignoni, P., Montani, C., & Scopigno, R. (1998). A comparison of mesh simplification algorithms. Computers & Graphics, 22(1), 37–54. http://doi.org/10.1016/S0097-8493(97)00082-4 Criminisi, A., Pérez, P., & Toyama, K. (2004). Region filling and object removal by exemplar-based image inpainting. IEEE Transactions on Image Processing, 13(9), 1200–1212. http://doi.org/10.1109/TIP.2004.833105 Crombez, N., Caron, G., & Mouaddib, E. (2015). 3D point cloud model colourization by dense registration of digital images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40- 5/W4, 123–130. http://doi.org/10.5194/isprsarchives-XL-5-W4-123-2015 Dall’Asta, E., Bruno, N., Bigliardi, G., Zerbi, A., & Roncella, R. (2016). Photogrammetric techniques for promotion of archaeological heritage: The archaeological museum of Parma (Italy). The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 41(B5), 243–250. http://doi.org/ 10.5194/isprs- archives-XLI-B5-243-2016 Díaz-Marín, C., Aura-Castro, E., Sánchez-Belenguer, C., Vendrell-Vidal, E., Abate, A. F., & Narducci, F. (2015). Virtual Reconstruction and Representation of an Archaeological Terracotta Statue. In 2nd Digital Heritage International Congress (DigitalHeritage) (pp. 699–702). Granada, Spain. http://doi.org/10.1109/DigitalHeritage.2015.7419602 Digne, J., Morel, J. M., Audfray, N., & Lartigue, C. (2010). High fidelity scan merging. Computer Graphics Forum, 29(5), 1643–1651. http://dx.doi.org/10.1111/j.1467-8659.2010.01773.x Happa, J., Williams, M., Turley, G., Earl, G., Dubla, P., Beale, G., & Chalmers, A. (2009). Virtual Relighting of a Roman Statue Head from Herculaneum: A Case Study. In 6th International Conference on Computer Graphics, Virtual Reality, Visualisation and Interaction in Africa (pp 5–12). Pretoria, South Africa. http://dx.doi.org/10.1145/1503454.1503456

Virtual Archaeology Review, 8(17): 49-55, 2017 54 DIGITIZATION OF RELIGIOUS ARTIFACTS WITH A STRUCTURED LIGHT SCANNER

Hermon, S., Iannone, G., Fakka, M., Khalaily, H., Avni, G., & Re’em, A. (2013). Digitizing the Holy - 3D Documentation and analysis of the architectural history of the ”Room of the Last Supper” - the Cenacle in Jerusalem. In 1st Digital Heritage International Congress (DigitalHeritage), (pp. 359–362). Marseille, France. http://dx.doi.org/10.1109/DigitalHeritage.2013.6744780 Kiourt, C., Koutsoudis, A., Arnaoutoglou, F., Petsa, G., Markantonatou, S., & Pavlidis, G. (2015). A dynamic web-based 3D virtual museum framework based on open data. In 2nd Digital Heritage International Congress (DigitalHeritage) (pp. 647–750). Granada, Spain. http://doi.org/10.1109/DigitalHeritage.2015.7419589 Lanitis, A., Stylianou, G., & Voutounos, C. (2012). Virtual restoration of faces appearing in byzantine icons. Journal of Cultural Heritage, 13(4), 404–412. http://dx.doi.org/10.1016/j.culher.2012.01.001 López-Moreno, J., Hadap, S., Reinhard, E., & Gutierrez, D (2010). Compositing images through light source detection. Computers & Graphics, 34(6), 698–707. http://dx.doi.org/10.1016/j.cag.2010.08.004 Luo, M., & Bors, A. G. (2011). Surface-preserving robust watermarking of 3-D shapes.IEEE Transactions on Image Processing, 20(10), 2813–2826. http://dx.doi.org/10.1109/TIP.2011.2142004 Menna, F., Nocerino, E., Remondino, F., Dellepiane, M., Callieri, M., & Scopigno, R. (2016). 3D digitization of an heritage masterpiece – A critical analysis on quality assessment. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 41(B5), 675–683. http://dx.doi.org/10.5194/isprsarchives-XLI-B5-675-2016 Nabil, M., & Saleh, F. (2014). 3D reconstruction from images for museum artefacts: A comparative study. In International Conference on Virtual Systems & Multimedia (VSMM), (pp. 257–260). Hong Kong. http://dx.doi.org/10.1109/VSMM.2014.7136681 Nicolae, C., Nocerino, E., Menna, F., & Remondino, F. (2014). Photogrammetry applied to problematic artefacts. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40(5), 451–456. http://dx.doi.org/10.5194/isprsarchives-XL-5-451-2014 Pérez, A., & Robleda, G. (2015). 3D Virtualization by close range photogrammetry indoor Gothic Church Apses. The case study of Church of San Francisco in Betanzos (La Coruña, Spain). The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40-5(W4), 201–206. http://dx.doi.org/10.5194/isprsarchives-XL-5-W4-201-2015 Polhemus (2016). FastSCAN scanner. Retrieved October 10, 2016, from http://polhemus.com/scanning- digitizing/fastscan/ Remondino, F. (2011). Heritage recording and 3D modeling with photogrammetry and 3D scanning. Remote Sensing, 3(12), 1104–1138. http://dx.doi.org/10.3390/rs3061104 Rinaudo, F., Chiabrando, F., Lingua, A., & Spanò, A. (2012). Archaelogical site monitoring: UAV photogrammetry can be an answer. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 39(B5), 583–588. http://dx.doi.org/10.5194/isprsarchives-XXXIX-B5-583-2012 Rodriguez-Gonzálvez, P., Nocerino, E., Menna, F., Minto, S., & Remondino, F. (2013). 3D surveying & modeling of underground passages in WWI fortifications. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40(5(W4)), 17–24. http://dx.doi.org/10.5194/isprsarchives-XL-5-W4-17-2015 Salimbeni, R., Pini, R., & Siano, S. (2003). The Optocantieri project: toward a synergy between optoelectronics and information science for cultural heritage conservation. In SPIE 5146 - The International Society for Optical Engineering, (pp. 24–33). http://dx.doi.org/10.1117/12.506181 Sander, P. V., Snyder, J., Gortler, S. J., & Hoppe, H. (2001). Texture mapping progressive meshes. In Proceedings of the 28th annual conference on Computer graphics and interactive techniques (SIGGRAPH '01). ACM, New York, NY, USA, (pp. 409–416). http://dx.doi.org/10.1145/383259.383307 Soria, G., Ortega, L., Feito, F. R., & Barroso, I. (2015). Processing panoramic images in heritage. Jaen Cathedral. Virtual Archaeology Review, 6(13), 28–34. http://doi.org/10.4995/var.2015.4368 Styliani, S., Fotis, L., Kostas, K., & Petros, P. (2009). Virtual museums, a survey and some issues for consideration. Journal of Cultural Heritage, 10(4), 520–528. http://dx.doi.org/10.1016/j.culher.2009.03.003 Van Gasteren, M. (2013). Fiel reproducción de Santa Teresa de Jesús y Cristo ”El Amarrado” de Gregorio Fernández a base de Tecnologías 3D. Virtual Archaeology Review, 4(8), 33–36. http://dx.doi.org/10.4995/var.2013.4284 Wang, K., Lavoué, G., Denis, F., & Baskurt, A. (2011). Robust and blind mesh watermarking based on volume moments. Computer & Graphics, 35(1), 1–19. http://dx.doi.org/10.1016/j.cag.2010.09.010

Virtual Archaeology Review, 8(17): 49-55, 2017 55

Virtual Archaeology Review, 8(17): 56-63, 2017 http://dx.doi.org/10.4995/var.2017.4726 © UPV, SEAV, 2015

Received: February 24, 2016 Accepted: November 30, 2016

DIGITAL HERITAGE TRAINING FOR HISTORIANS IN EUROPE: A LOCAL PROPOSAL APRENDIZAJE DIGITAL EN PATRIMONIO PARA HISTORIADORES EUROPEOS: UNA PROPUESTA LOCAL José-Antonio Ruiz Gil Department of History, Geography and Philosophy, University of Cadiz, Avda. Gomez Ulla s/n, 11500 Cadiz, Spain. [email protected]

Abstract: This paper presents a proposal for the use of digital resources to improve the university curriculum for historians in particular and heritage managers in general. It is possible to develop the sector, providing more employment and promoting theoretical and methodological changes conducive to mutual progress. To achieve this, the proposal takes into account the recent legal reforms in education and within the European Digital Single Market. However, although such changes are possible given that there is already extensive experience in heritage digitisation, they are hindered by the current curricula content and slow implementation of competency-based education. The desk research reported here focused on a Spanish context that could benefit from curriculum development implemented elsewhere, and the resulting proposal for positive action was explored in the context of history and heritage education at the University of Cadiz. Key words: digital training, historian, cultural heritage, European Higher Education Area (EHEA).

Resumen: Este artículo propone implementar el uso de los recursos digitales en el currículo de historiadores en particular, y de gestores del patrimonio en general. Es posible desarrollar el sector proporcionando más empleo y promoviendo cambios teóricos y metodológicos. Para alcanzar esta propuesta hay que tener en cuenta las recientes reformas legales en educación y en el Mercado Europeo Único Digital. Tales cambios son posibles porque hay una enorme experiencia en la digitalización del patrimonio. Pero los contenidos de los planes de estudio existentes en la actualidad, así como la lentitud en la implantación de la Educación por Competencias lo impiden. Esta investigación documental se basa en un contexto español que puede beneficiarse del desarrollo curricular implementado en otros lugares. Se plantea todo esto en el caso de los estudios de Historia y Patrimonio en la Universidad de Cádiz, así como una propuesta de acción favorable. Palabras clave: aprendizaje digital, historiador, patrimonio cultural, Área de Educación Superior en Europa (EHEA).

The humanities have been slow to incorporate the new 1. Introduction technologies, as noted by Pedro Peña (2015), who in The 2015 Digital Heritage Congress represents a particular highlighted the reasons for conservatism and benchmark for all those in academia who work in the disconnection from the labour market and business humanities. In my opinion, this is due to the crisis innovations. The work of Roberto Busa with IBM in 1949 currently afflicting the humanities. This Congress is not a (Hayles, 2012, note 1 in page 62; Jones, 2016) is also discipline-based discussion forum, it far transcends that. important in this respect. Two works in particular have It assumes that the humanities are immersed in a new emphasised the economic importance and relevance of socio-historical environment. new technologies in the field of culture in general, and of heritage in particular. In the case of culture, the book Here, I want to use the term 'historian' in a broad sense, Creation of Enterprises in the Field of Culture, directed to refer to professionals engaged not only in history but by José Ruiz Navarro from the University of Cadiz, is a also in heritage. I focus here on heritage professionals good example. Originally written in 2005 (and published because it is precisely on their metier that I wish to in Ruiz, 2008), this was a groundbreaking study of how reflect. the new technologies drive start-ups and help redress the principle established by Baumol and Bowen (it This line of inquiry obviously reflects not only a way of involves a rise of salaries in jobs that have experienced working but also an inner conviction that I think is widely no increase of labour productivity). The COTEC Report accepted: namely that history is constructed from (2010) focused on innovation in the field of historical material traces that, with the corresponding social heritage, and mentioned the importance of technology conceptualisations, we now call historical or cultural applied to heritage in several fields. heritage. Therefore, I will discuss historians as heritage professionals, and I intend to talk about their training in In the francophone world, the digital humanities are relation to the goal of this Congress. known as the “humanités numériques” (Barbier, 2014),

*Corresponding author: José-Antonio Ruiz Gil, [email protected] 56

DIGITAL HERITAGE TRAINING FOR HISTORIANS IN EUROPE: A LOCAL PROPOSAL and have formed a part of the curriculum for some years We cannot practice our profession without considering now at institutions such as the University of Tours the presence of historiographical options that rely on the (Lorans, 2008). existence of the Internet (such as History under Debate), or which attempt to address humanistic research from a The most important issue is the future of digital scientific perspective, as Lev Manovich proposed when competencies. In the academic year 2015-2016, the new he coined the term Cultural Analytics in 2005 (Manovich, primary school curriculum introduced the subject “Digital 2016), and subsequently gave it a more practical form Culture and Practice” (Regional Government of two years later. In a more business-oriented sphere, the Andalusia, JUNTA, 2015) in sixth grade (11 years old concept of culturomics has already been proposed as an children) within the context Autonomous Region powers approach to working with vast amounts of data (for established by the LOMCE (Education Act), which example, all books or all manuscripts, etc.) (Michel et al., accounts for 7% of the total teaching hours (two a week, 2011). In short, historians must embrace this context or Social Sciences and Natural Sciences). now that knowledge is digitally produced and shared. The course is structured into three thematic blocks: Digital Culture, Technological Practice and Online 3. From European to local scene Education. The first of these is intended to introduce students to the importance of digital technology in Earlier, I referred to the existence of the EHEA society and the media (social networking platforms, (European Higher Education Area) as a frame of blogs, mobile connectivity), and digital identity (use, reference for university education. This obliges us to safety and ethics). The second explores mobile and implement the proposal made on the 6th of May, 2015, tablet applications, as well as blogs, wikis, virtual by the President of the European Union, Mr Juncker, communities and social networks (web 2.0), and the concerning the European Digital Single Market third looks at digital learning environments and the (European Commission, 2015). This proposal should be production of original work. given priority at the highest political, economic and social level. It revolves around three ‘pillars’ or areas of political 2. Why do we need digital historical action. heritage? The first aim is to improve online access to digital goods and services in the European Union, improved access In general, we define history as information about the being understood here mainly in terms of modernising past. When this history has a wider scope and a material copyright legislation regarding images, films, music and expression, we call it heritage. Given that we now live in games. At this point, I would like to draw the audience’s the information age, how does this influence heritage? attention to the historical and cultural importance of From my point of view, information from the past some of these fields. Heritage is primarily about image. expressed in the analogue form, in other words, as a How can digital media help us to enrich heritage? But we continuous input signal, can be converted into a series of must also ask what training students of either history or discrete values. These numerical values, or digits, are cultural heritage have in digital methods, as presumably represented in binary code by computers. they have acquired considerable knowledge of history and cultural heritage – and depending on their institution Our everyday lives are filled with computers. The perhaps of digital methods pertaining to both fields. question is whether these are more than mere machines: can they change our understanding of The second aim is to create an environment in which history? Can heritage be digitally interpreted? The digital networks and services can flourish, in other answer is clear from the moment we become words, one in which they can improve in quality and participants in this Congress. We are not talking about increase in quantity. If this happens, and here I mean as the mere storage of information, or of a machine that can an everyday reality, it will become increasingly transmit narratives. When physical (normally visual) necessary for our students to acquire more digital skills. variables are converted, they are analogically expressed In short, proficient digital literacy is vital for all. by means of digital representations. This facilitates their processing (digitisation), by reducing noise and The third aim is for digital technology to become a interference. vehicle for economic and social growth. This essentially implies working with big data and cloud computing. I Information, in this case, historical information, is repeat, we need to increase our digital skills. embedded in a context, and our current context is digital. The very concept of ‘Digital Heritage’ proposed by this The digital competencies have been acknowledged as Congress is simply one part of the ‘digital humanities’. one of the 8 key competencies for lifelong learning by Digital media and computational analytics have already the European Parliament (2006) and the European been employed for several years now in the practice of Commission (2006). The digital competence was defined history, and by extension, of culture. Similarly, other in the DIGCOMP Project of the Institute for Prospective tools such as digital files, databases, presentations, Technological Studies (IPTS), one of the seven that form online access, interactive images, time lines, audio files, the European Joint Research Centre, as “the confident, Geographic Information Systems (GIS), 3D modelling critical and creative use of ICTs (Information and and virtual recreation, have for years accompanied the Communications Technologies) to achieve goals related analysis of big data, which are basically used in to work, employability, inclusion and/or participation in quantitative history, cliometrics and social computing society” (European Parliament, 2006 & European (Grandjean, 2015). A recent example of the evaluation of Commission, 2006). digital programmes implemented in museums and Digital competence is a key cross-curricular competence exhibitions has been published in this journal (Vicent, involved in the acquisition of basic skills such as Ibáñez-Etxeberria and Asensio, 2014). language, maths, learning to learn and creativity, as

RUIZ GIL, 2017 explicitly described in European legislation content such as GIS and photogrammetry is included in (COMMUNICATION, 2016). archaeology. In Spain, the main areas of digital competence identified in a document issued by the Spanish Ministry 4. Comparing the curricula in Spain of Education, Science and Sport (MINISTERIO, 2013) In The University of Cadiz is no exception in the Spanish are: university system. To explain this, I will conduct a - Information: to identify, retrieve, store, organise and comparison of the curricula of other universities with analyse digital information, evaluating its purpose degrees related to history. It is not easy to establish a and relevance. comparison between the different curricula. The first problem encountered stems from the lack of definition of - Communication in digital environments: areas of knowledge, as some are historically collaborative learning, interaction and participation in established, whereas others have traditionally been communities and networks. included within the arts and humanities. Thus, the - Creation of multimedia and computer programming Autonomous University of Barcelona (UAB, 2016) offers content (text, images, videos), integrating and a Master’s Degree in Digital Humanities, in which the reformulating knowledge within a framework of curriculum focuses on philology, a trend that is common intellectual property rights and licences. in the humanities in Spain. The subjects explicitly mentioning the word ’digital’ are Principles and Methods - Security: data protection and digital identity. of Digital Humanities, Speech and the Digital Universe, - Troubleshooting and identification of needs and and Digital Philology. Another example of this trend is digital resources, creative use and updating. the Master’s Degree in Digital Humanities and Textual Heritage at the University of Salamanca (USAL, 2016). Digital skills are becoming of fundamental importance. In this case, I will discuss the digital textualities subjects, However, they do not explicitly appear in the Spanish Creation, Transmission and Dissemination of Digital Qualifications Framework for Higher Education (in Contents, and the optional Digital Reading and Writing: Spanish, the MECES), unless we consider them to be Devices and Applications. Thus, to sum up, this will not encompassed within “learning skills”, and practical skills be our field of analysis, since furthermore it is associated such as “Knowledge of information technology related to with the International Society of Hispanic Digital the field of study”. Humanities, which in 2013 organised the First Conference on Digital Humanities. Challenges, In the case of the University of Cadiz, there is only one Achievements and Future Prospects (HDH, 2013). ‘digital’ Master’s programme, the Master's Degree in Digital and Social Marketing. Furthermore, the digital Turning to culture, for example the Master’s Degree in age is completely absent in the content of courses Cultural Management at the Carlos III University of (undergraduate and master’s degrees) related to Madrid, Module 6 (Culture and Technology) includes the history and heritage. As we are discussing skills, I will subject Technology Applied to Work in Culture and illustrate the specific competence level in history Current Cultural Media, and Module 7 (Heritage) (University of Cadiz, 2008): includes Application of Technology to Cultural Heritage G05 - To acquire basic skills in document (UC3M, 2016). Similarly, there is an Advance Research management and analysis for the production of and Innovation Applied to Heritage subject on the historical and cultural heritage knowledge. Master’s Degree in Cultural Heritage: Research and Management, taught at the University of Castile-La G09 - To be proficient in the activities and planning Mancha (UCLM, 2016), while the Master’s Degree in entailed in the cultural and publishing market and Historic, Artistic and Cultural Research and Managenent industry. at the University of Murcia includes a research specialisation with the compulsory subject, Advanced E02 - To know how to search and manage Research Techniques and Methods (UM, 2016). This bibliographic resources. superficial approach is also evident in the Master’s E03 - To know how to analyse and interpret diverse Degree in World Heritage and Cultural Projects for historical sources. Development, which the universities of Barcelona, Turin and the International Training Centre of the ILO in E06 - To know how to analyse and interpret the collaboration with the UNESCO Centre for World archaeological record. Heritage have entitled: ICT Applied to Cultural and Any of these skills could have a digital component, Heritage Projects (ITCILO, 2016). such as reading, writing, analysing and or interpreting Let us return to heritage, and more specifically, to digital data: the development of Bloom’s Digital conservation. The Master’s Degree at the Pablo de Taxonomy from its analogue version, Bloom’s Olavide University in Seville includes the optional Taxonomy, could be seen as an example of this subject, New Imaging Technologies Applied to Cultural progression (Fractus Learning, 2014). Heritage (GIS and Image), while the Master’s Degree in As can be seen, traditional skills in the Master’s Degree Arts, Museums and Historical Heritage Management at in Heritage, Archaeology and Maritime History adhere to the same university includes Digital Museology: The the EC 9 ability to identify ICT applications in heritage, New Museums (UPO, 2016). However, this so-called especially multimedia, television and radio resources. ‘digital’ content does not refer to the application of digital These competences do not seem adequate to enter the knowledge, but to the restoration of digital material. This European labour market, al least not in the form in which same problem is evident in Subject 4, Conservation they have been designed in the European Union. Strategies for Digital Heritage, on the Master’s Degree in Nonetheless, closer inpection indicates that basic digital Conservation of Cultural Heritage taught at the

Virtual Archaeology Review, 8(17): 56-63, 2017 58 DIGITAL HERITAGE TRAINING FOR HISTORIANS IN EUROPE: A LOCAL PROPOSAL

Complutense University of Madrid (UCM, 2016), and the The Department of Graphic and Cartographic University of Malaga’s independent course (not a Expression at the University of Alicante, in conjunction master’s degree) on History of Digital Art with the Virtual Heritage Group, has designed a specific (CEHAD, 2016). This becomes even more specific when Master’s Degree in Virtual Heritage, which includes a considering the conservation of architectural heritage. In specialisation course on New Technologies Applied to the case of the master’s degree with the same title Virtual Heritage, as well as a specialist course on Virtual taught at the Polytechnic University of Valencia, the Restoration and an expert course on Heritage word ‘digital’ is not used, but the subjects clearly include Virtualisation. This has a solid tradition in architecture digital content: Advanced Technology Applied to the and engineering, but has not yet been consolidated in Conservation of Architectural Heritage, Advanced officially recognised degrees (UA, 2016). The full text Techniques for Elevations: Topography, can be accessed in this journal (Molina, Esclapés, Photogrammetry and Laser Scanner, and Tejerina, & Fabregat, 2013). Virtual Reconstruction of Architectural Heritage (UPV, 2016a). To end the comparison, I will refer to the most important training offered, although it has not yet achieved officially Continuing my argument, at the same university (the recognised status in higher education, namely the Polytechnic University of Valencia), the Master’s Degree Master’s Degree and Expert Course in Archaeology and in Cultural Management only includes the subject Virtual Heritage: Documentation, Preservation and “Knowledge Management and Information Resources” Dissemination of Heritage in the Digital Age. This forms (UPV, 2016b), but the Master's Degree in Cultural part of the International Campus in Archaeology and Heritage: Identification, Analysis and Management at the Heritage SEAV Virtual Training, a space created by the University of Valencia includes the subjects Spanish Society of Virtual Archaeology, the Virtual Historiography and New Technologies Applied to Archaeology International Network, INNOVA, and 18 Cultural Heritage and the specialisation in Analysis and research groups affiliated to 14 universities (SEAV, Management of Landscape Heritage, “Landscape 2012). From my point of view, it is appropriate to Analysis Techniques”, features Landscape Mapping mention the six professional areas that underpin the Techniques, Thematic mapping and GIS, Aerial Master’s Degree: photography and Satellite Imagery, and Graphic and Photographic Representation Techniques (UV, 2016). - Methodological Innovation in Heritage. However, it is in archaeology where the presence of the - Advanced Heritage Visualisation. digital age is most clearly evident. The University of - Interpretation and Presentation of Heritage. Alcalá de Henares offers an independent expert course in the arts and humanities which includes the subjects - Geometric Documentation of Heritage. Digital Tools Applied to Archaeology, and entails use of - Virtual Reconstruction of Heritage. GIS tools, the use of GIS in archaeology and digital drawing: archaeological/planimetric materials - 3D Heritage Research. (UAH, 2016). Turning to the official Master’s Degree in Classical Archaeology, promoted by a consortium of 5. Discussion Catalan universities and the Catalan Institute of Classical Archaeology (ICAC), this merely includes In 2008, the Excellence in Processing Open Cultural Drawing and Restoration Techniques in Architecture, Heritage group published a draft proposal for a digital and Landscape Archaeology Methods and Techniques heritage curriculum (EPOCH, 2008) aimed at a number (URV, 2016). Similarly, the Master’s Degree in of graduate programmes in Italy, the UK, Greece, Mediterranean Archaeology in Classical Antiquity, taught France and Spain (Farjas and Rejas, 2008). My analysis at the Complutense University of Madrid (UCM DSU, thus represents the latest contribution on the reality of 2016), has a course on GIS and another on digital heritage in Spain. Documentary Sources and Resources in Digital Despite educational reforms and the existence of the Archaeology. EHEA, historians receive a university education that is The curricula at the universities of Granada, Jaen and obsolete and anachronistic. I believe that this is basically Seville notably include an undergraduate course on New due to an illogical emphasis in the curriculum Technologies for the Dissemination and Enhancement of (excessively political and lacking in social awareness, Archaeological Heritage (UGRA, 2016). This indicates a understood in terms of work experience and international greater and more specific use of these technologies, scope), as well as to the enduring conservatism inherent especially compared to other, similar degree courses, to the academic structure. I do not intend to claim that no which employ more traditional approaches. Moreover, Spanish historians on any of these programmes are there is a growing interest in specialisation and engaged in making digital knowledge work. But these retraining, evidenced by the fact that several courses programmes are insufficient to achieve the goals and seminars have already been scheduled on master’s proposed in the EU. Furthermore, time is running out as degrees: the courses include Instrumental Aspects other educational and social sectors move more rapidly, Related to Archaeological Exploration and Excavation, demonstrating the lack of inter/cross disciplinarity. Geophysics Applied to Archaeology, Quantitative Digital heritage curricula are rare in Europe. Exceptions Methods and Computer Science Applied to Archaeology, include the Digital Art History Initiative of the Getty while the seminars include GIS Applications in Foundation, a series of training workshops funded via Archaeology, 3D Modellind and Reconstruction with grants awarded to trainee art historians to work with Laser Scanners in Archaeological and Historical digital technologies (Getty, 2016). This initiative is based Heritage (UGRA-MASTER, 2016). on “Transitioning to a Digital World. Art History, its

RUIZ GIL, 2017

Research Centres and Digital Scholarship. A report to 6. Conclusion the ‘The Samuel H. Kress Foundation and Roy Rosenzweig Centre for History and New Media, George My proposal is not novel in the context of the last Mason University’” (Zorich, 2012). International Digital Heritage Congress, where the Master´s Degree in Virtual Cultural Heritage was These institutions came together again in 2014 to announced. It is far from my intention to underestimate organise the summer activity, Rebuilding the Portfolio: or dismiss the digital work that has been carried out Digital Humanities for Art Historians (George Mason throughout history. Rather, I have tried to highlight the Univ., 2014). Also for the summer, in this case for 2017, slow pace at which it is implemented. The objective a course on 3D Laser Scanning and Photogrammetry underlying my proposal is not solely to achieve has been announced, in this case in Nafplio, Greece specialisation in digital material, but also to incorporate (IHC, 2016). digital skills in the standard curriculum for historians and However, the growing importance of digital heritage heritage professionals, or “heritagians”. training is evidenced not only by the organisation of The opportunity and desirability of the new European seasonal courses, but also by its implementation in Digital Single Market requires us to plan courses British universities. I cite as an example the universities consistent with a digital future. In my view, a proposal for of Manchester and Leicester. The former offers a Digital a more technologically focused training is logical; not for Heritage Research Training Initiative, with a course the market as such, but rather for a society that stopped worth 15 credits on Digital Heritage, within the area of being analogue some time ago. In this regard, mention Digital Training, which engages with issues related to should be made of the work in Europe undertaken by the curation, interpretation, communication and learning in Initial Training Network for Digital Cultural Heritage, heritage institutions (Manchester Univ., 2016). which emerged from the FP7 PEOPLE programme In the latter, courses on Heritage Practice Training are coordinated by the Digital Heritage Research Lab of the offered in conjunction with the Historic England and Cyprus University of Technology (ITN-DCH, 2016), and Heritage Skills Centre (Leicester Univ., 2016). of the Heritage Portal, which provides general information on training opportunities in Europe The training of professionals in the humanities in (Heritage Portal, 2016). general, and in heritage and history in particular, cannot be based solely on traditional content or methodologies. For history and heritage studies at the University of Taking this as an axiom in this paper, history is alredy Cadiz, it will be necessary to incorporate new material to being practised differently from a quantitative and expand the digital skills of students graduating within the qualitative point of view (Turchin et al., 2013; framework of the knowledge society. Such material EPOCH, 2008). should cover not only the creation of digital products, but also the minimum digital skills of knowledge and use of Material on virtual reality is now being taught on various hardware and software related to GIS, georeferencing, specialised courses, even at master’s degree level. remote sensing, LiDAR, virtualisation, photogrammetry, Virtualisation is perhaps the most important part of 3D scans, and surveying, drawing and recording digitalising the humanities, as I argued in section 4, but archaeological excavations. precisely for this reason there is an evident need to include basic training. At university level, we must There is no time to lose. Our responsibility is to prepare integrate and promote the different practical activities future history and heritage professionals, precisely those within the new digital universe that history will shortly who will replace us in a very different society as regards invite us to participate in. education. This is the path that the authors and institutions I mentioned in Sections 2 and 3 have When we compare the Gutenberg Galaxy (McLuhan proposed we follow. It is not about blindy taking a path 1962), based on the alphabet, we move from a debate but about appreciating how far we have come and how focused on encoding and decoding (reading and writing) far we have to go. But are we convinced that this is the to another which requires skill, knowledge and aptitudes. right direction? I think so. The acquisition of competencies involves acting and interacting (agency), with technology, learning its uses Acknowledgements for good professional practice. Again, we should recall This work was supported by the CEI PatrimoniUn10, the Bloom’s Taxonomy and its application to digital contexts Seminario ‘Agustín de Horozco’, and the Grupo de (Fractus Learning, 2014). For a deeper exploration of Estudios de Historia Actual (GEHA) of the University of this line of research, I recommend reading the recent Cadiz. work by David Barreiro (2013).

References Barbier, B. (2014). Les offres de formation en patrimoines et humanités numériques dans le établisements d’enseignement supérieur française. In Bernadette Saou-Dufrene (Ed.), Heritage and digital humanities. How should training practices evolve? (pp. 395–399). Berlin: Lit Verlag. Barreiro, D. (2013). Arqueológicas. Hacia una arqueología aplicada. Barcelona: Bellaterra Arqueología. CEHAD (2016). Curso de extensión sobre historia del arte digital. Retrieved October 12, 2016, from http://www.cehad.es/

Virtual Archaeology Review, 8(17): 56-63, 2017 60 DIGITAL HERITAGE TRAINING FOR HISTORIANS IN EUROPE: A LOCAL PROPOSAL

Communication from the Commission to the European Parliament. The Council, the European Economic and Social Committee and the Committee of the Regions, 2016 . Digitising European industry reaping the full benefits of a digital single market. Brussels, 19.4.2016 COM, 180 final. Retrieved from http://eur-lex.europa.eu/legal- content/EN/TXT/?uri=CELEX%3A52016DC0180 COTEC (2010). Informes sobre el sistema español de innovación en el sector del patrimonio histórico. Madrid: Fundación COTEC para la Innovación Tecnológica. EPOCH (2008). Draft proposal for a curriculum for digital heritage studies. Policies, practices and developments in Europe (pp. 37–47). Retrieved October 12, 2016, from http://public-repository.epoch- net.org/deliverables/D_4_13_Final_State_of_the_Union.pdf European Commission (2015). Digital single market. Retrieved February 19, 2016, from http://ec.europa.eu/priorities/digital-single-market/ European Parliament (2006). Recommendation 2006/962/CE. & European Commission, 2006. DOCE L394, 30/12/2006. Retrieved February 20, 2016, from http://is.jrc.ec.europa.eu/pages/EAP/DIGCOMP.html

Farjas, M. & Rejas, G. (2008). Digital cultural heritage-the Spanish reality. In EPOCH Draft proposal for a curriculum for digital heritage studies. Policies, practices and developments in Europe (pp. 21–36). Retrieved from http://public- repository.epoch-net.org/deliverables/D_4_13_Final_State_of_the_Union.pdf Fractus Learning (2014). Bloom’s digital taxonomy. Retrieved June 6, 2016, from http://www.fractuslearning.com/wp- content/uploads/2014/08/bloom_taxonomy_1280.png George Mason University (2014). Rebuilding the portfolio: Digital humanities for art historians. Summer Institute George Mason University, July 7-18. Retrieved October 12, 2016, from http://arthistory2014.doingdh.org/wp-content/uploads/ sites/3/2014/01/ReBuildingthePortfolioCurriculum2014.pdf Getty (2016). Digital art history initiative of the Getty Foundation, a whole of training workshops supporting with awarded grants for preparing art historians to work with digital technologies. Retrieved October 12, 2016, from http://getty.edu/foundation/initiatives/current/dah/index.html Grandjean, M. (2015). La visualisation des données en Histoire. Histoire et Informatique/ Geschichte und Informatik, vol. 18/19, 109–128. Retrieved from www.chronn-verlag.ch/php/chronos_src/Buch/978-3-0340-1289-8_Buch.pdf

Hayles, N. K. (2012). How we think: Transforming power and digital technologies. In D. Berry (Ed.): Understanding digital humanities, Palgrave MacMillan (pp. 42–66). http://doi.org/10.1057/9780230371934 HDH (2013). Humanidades Digitales Hispánicas Congreso Internacional. Retrieved February 21, 2016, from http://hdh2013.humanidadesdigitales.org/ Heritage Portal (2016). Retrieved October 12, 2016, from http://www.heritageportal.eu/Resources/Training/ IHC (2016). Initiative for Heritage Conservation. Retrieved October 12, 2016, from http://www.inherity.org/training/summer-schools/digital-tools-for-heritage-management/ ITCILO (2016). Máster en Patrimonio de la Humanidad y Proyectos culturales para el desarrollo. Retrieved February 21, 2016, from http://patrimoniomundial.itcilo.org/programa-1/copy_of_programa-1 ITN-DCH (2016). Initial training network for digital cultural heritage: Projecting our past to the future. Retrieved October 12, 2016, from http://www.itn-dch.eu/index.php/about-us/

Jones, S. E. (2016). Roberto Busa, S. J., and the emergence of humanities computing: The Priest and the punched cards. Routledge. http://dx.doi.org/10.4324/9781315643618 Junta de Andalucia (2015). Orden de 17 de marzo de 2015, por la que se desarrolla el currículo correspondiente a la Educación Primaria en Andalucía. Retrieved February 19, 2016, from http://www.juntadeandalucia.es/eboja/2015/60/BOJA15-060-00134-5243-01_00066439.pdf Leicester University (2016). Heritage practice training courses of the Leicester University. Retrieved October 12, 2016, from http://www2.le.ac.uk/departments/history/heritage

Lorans, Elisabeth (2008). Digital archaeology curricula: some aspects of the state os practice in France and Tours University experiences. In EPOCH Draft proposal for a Curriculum for Digital Heritage Studies. Policies, Practices and Developments in Europe (pp. 37–47). Retrieved from http://public-repository.epoch- net.org/deliverables/D_4_13_Final_State_of_the_Union.pdf

Virtual Archaeology Review, 8(17): 56-63, 2017 61 RUIZ GIL, 2017

Manchester University (2016). Digital heritage research training initiative (2008-2010). Retrieved October 12, 2016, from http://www.alc.manchester.ac.uk/icp/research/projects/digital-heritage-training/ Manovich, L. (2016). The science of culture? Social computing, digital humanities and cultural analytics. In Mirko Tobias Schaefer & Karin Van Es (Eds.), The datafield society. social research in the age of Big Data. Amsterdam: Univ. Press. Retrieved from http://manovich.net/index.php/projects/cultural-analytics-social-computing McLuhan, M., 1962. The Gutenberg galaxy: the making of tipographic man. Univ. Toronto Press. Michel, J.-B., Y.K. Shen, A. P. Aiden, A. Veres, M. K. Grey, The Google Books Team, J.P. Pickett, D. Holberg, D. Clancy, P. Norvig, G. J. Orwant, S. Pinjer, M. A. Nowak, & Lieberman Aiden, E. (2011). Quantitative analysis of culture using millions of digital books. Science, 331,(6014), 176–182. http://dx.doi.org/10.1126/science.1199644 Ministerio de Educación, Ciencia y Deporte (2013). Marco común de competencia digital docente, Instituto Nacional de Tecnologías Educativas y de Formación del Profesorado, V 2.0. Retrieved February 20, 2016, from http://educalab.es/documents/10180/12809/MarcoComunCompeDigiDoceV2.pdf Molina Vidal, J., Esclapes Jover, F.J., Tejerina Anton, D., & Fabregat Bolufer, L. (2013). Virtualización del Patrimonio de la Universidad de Alicante. Propuesta formativa. Virtual Archaeology Review, 4(8), 100–103. http://dx.doi.org/10.4995/var.2013.4331 Peña Dominguez, P. (2015). ¿Es la reconstrucción virtual de patrimonio infoarquitectura? In Congreso Internacional de Cultura Digital. Sociedad y Comunicación, perspectivas en el siglo XXI. 16-17 de marzo de 2015 (pp. 515–530). Zaragoza: Ediciones Universidad San Jorge. Ruiz Navarro, J. (Dir.) (2008). Creación de empresas en el ámbito de la Cultura. Madrid: Datautor. SEAV (2012). Máster y Experto en Arqueología y Patrimonio Virtual. Retrieved February 21, 2016, from http://blocs.gencat.cat/blocs/AppPHP/tribunadarqueologia/files/2013/01/PDF-MASTER-Y-EXPERTO-SEAV.pdf Turchin, P., Currie, Th. E., Turner, E. A. L., & Gravilets, S. (2013). War, space, and the evolution of old world complex societies, PNAS, 110(41), 16384–16389. http://dx.doi.org/10.1073/pnas.1308825110 UA (2016). Máster título propio de la Universidad de Alicante en Patrimonio Virtual. Retrieved February 21, 2016, from http://www.patrimoniovirtual.com/

UAB (2016). Universidad de Barcelona Máster en Humanidades Digitales. Retrieved February 21, 2016, from http://www.uab.cat/web/postgrado/master-en-humanidades-digitales/informacion-general-1206597472083.html/ param1-3202_es/param2-2006/ UAH (2016). Título de Experto en Artes y Humanidades por la Universidad de Alcalá de Henares. Retrieved February 21, 2016, from http://www.universia.es/estudios/uah/experto-herramientas-digitales-aplicadas-arqueologia/st/242248 UC3M (2016). Máster en Gestión Cultural de la Universidad Carlos III de Madrid. Retrieved February 21, 2016, from http://www.mastergestioncultural.eu/ UCLM (2016). Máster universitario en Patrimonio Histórico: investigación y gestión, de la Universidad de Castilla-La Mancha. Retrieved February 21, 2016, from https://muphig.masteruniversitario.uclm.es/plan.aspx UCM (2016). Máster universitario en Conservación del Patrimonio Cultural de la Universidad Complutense de Madrid. Retrieved February 21, 2016, from https://www.ucm.es/masterconservacion/estructura UCM DSU (2016). Máster Universitario en Arqueología del Mediterráneo en la Antigüedad Clásica, de la Complutense de Madrid. Retrieved February 21, 2016, from https://www.ucm.es/master-arqueologia-antiguedad/estructura-del- plan UGRA (2016). Grado Interuniversitario en Arqueología. Retrieved February 21, 2016, from http://grados.ugr.es/arqueologia/pages/infoacademica/estudios UGRA-Master (2016). Máster Interuniversitario en Arqueología. Retrieved February 21, 2016, from http://masteres.ugr.es/arqueologia/pages/ficha UM (2016). Máster Universitario en Investigación y Gestión del Patrimonio Histórico-Artístico y Cultural por la Universidad de Murcia. Retrieved February 21, 2016, from http://www.um.es/web/letras/contenido/ estudios/masteres/arte Universidad de Cadiz (2009). Memoria del Grado en Historia. Retrieved February 19, 2016, from http://filosofia.uca.es/filosofiayletras/portal.do?TR=A&IDR=1&identificador=1126

Virtual Archaeology Review, 8(17): 56-63, 2017 62 DIGITAL HERITAGE TRAINING FOR HISTORIANS IN EUROPE: A LOCAL PROPOSAL

UPO (2016). Máster de Arte, Museos y Gestión del Patrimonio Histórico, and Máster en Diagnóstico en Conservación del Patrimonio Histórico. Retrieved February 21, 2016, from https://www.upo.es/postgrado/Master-Oficial- Diagnostico-del-Estado-de-Conservacion-del-Patrimonio-Historico?opcion=3 & https://www.upo.es/postgrado/Master- Oficial-Arte-Museos-y-Gestion-del-Patrimonio-Historico?opcion=3 UPV (2016a). Universitat Politècnica de València Máster en Conservación del Patrimonio Arquitectónico. Retrieved February 21, 2016, from https://www.upv.es/titulaciones/MUCPA/menu_920396c.html UPV (2016b). Universitat Politêcnica de València Master Universitario en Gestión Cultural. Retrieved June 6, 2016, from https://www.upv.es/titulaciones/MUGC/indexc.html USAL (2016). Máster Universitario en Patrimonio Textual y Humanidades Digitales por la Universidad de Salamanca. Retrieved February 21, 2016, from http://www.usal.es/webusal/node/48570 URV (2016). Màster oficial universitari en Arqueologia Clàssica. Retrieved February 21, 2016, from http://www.urv.cat/masters_oficials/arts_humanitats/arqueologia_classica/es_pla_master_arqueologia_classica.html UV (2016). Master Universitario en Patrimonio Cultural: Identificación, análisis y gestión. Universitat de València. Retrieved June 6, 2016, from http://www.uv.es/uvweb/master-patrimonio-cultural-identificacion-analisis- gestion/es/master-patrimonio-cultural-identificacion-analisis-gestion-1285932165134.html Vicent, N., Ibañez-Etxeberria, A., & Asensio, M. (2014). Evaluation of heritage education technology-based programs. Virtual Archaeology Review, 6(13), 20–27. http://dx.doi.org/10.4995/var.2015.4367 Zorich, D. M. (2012). Transitioning to a digital world. art history, its research centers, and digital scholarship. A report to the The Samuel H. Kress Foundation and Roy Rosenzweig Center for History and New Media, George Mason University. Retrieved October 12, 2016, from http://www.kressfoundation.org/uploadedFiles/Sponsored_Research/ Research/Zorich_TransitioningDigitalWorld.pdf

Virtual Archaeology Review, 8(17): 56-63, 2017 63

Virtual Archaeology Review, 8(17): 64-74, 2017 http://dx.doi.org/10.4995/var.2017.6820 © UPV, SEAV, 2015

Received: November 11, 2016 Accepted: June 5, 2017

USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA) COMBINED USE OF DIGITAL NIGHTLIGHT PHOTOGRAPHY AND PHOTOGRAMMETRY IN THE PROCESS OF PETROGLYPHS DOCUMENTATION: THE CASE OF ALCÁZAR DE SAN JUAN (CIUDAD REAL, SPAIN) Víctor Manuel López-Menchero Bendichoa,*, Ángel Marchante Ortegab, Matthew L. Vincentc, Ángel Javier Cárdenas Martín-Buitragoa, Jorge Onrubia Pintadoa

a Universidad de Castilla-La Mancha, Instituto de Desarrollo Regional, Laboratorio de Arqueología, Patrimonio y Tecnologías Emergentes (LAPTE), Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain. [email protected]; [email protected]; [email protected] b G.I. Tarha, Departamento de Ciencias Históricas, Universidad de Las Palmas de Gran Canaria, c/ Pérez del Toro, s/n, 35003 Las Palmas de Gran Canaria, Spain. [email protected] C Universidad de Murcia, ITN-DCH Fellow, Centro de Estudios de Arqueología Virtual, Campus de Espinardo, 30001 Murcia, Spain. [email protected]

Abstract: For decades, the documentation of rock art has been the backbone of the research on this cultural practice. However, traditional techniques used for this purpose have proved to be imprecise and subjective. With the advent of the digital age there has been a revolution in the field of rock art documentation in general and in particular for petroglyphs. Techniques such as digital nightlight photography or three-dimensional (3D) photogrammetry have opened a world of possibilities. In the case of nightlight photography, its use goes back to the second half of the 20th century. However, in recent years the emergence of high-power digital cameras along with new lighting systems unthinkable decades ago, such as tactical flashlights, have rejuvenated a technique that seems incombustible. After the numerous tests carried out by the DIPAR (Integral Rock Art Documentation System) project team, it has become evident that, correctly employed, this technique still has immense potential in the field of rock art documentation, especially if we take into consideration its low cost and ease of use. Photogrammetry, on the other hand, is shown as the perfect complement to the records obtained through digital nightlight photography. The possibility of applying filters or shaders that increase the visibility of the digitized petroglyphs, within a metric scheme, is a great advance in their documentation. This paper explores the advantages associated with these techniques with a case study: unpublished petroglyphs of Alcázar de San Juan (Spain). Key words: digital photography, 3D, archaeological heritage, rock art, photogrammetry, documentation

Resumen: Durante décadas la documentación del arte rupestre ha constituido la columna vertebral de las investigaciones relativas a esta manifestación cultural. No obstante, las técnicas tradicionales empleadas para ello han demostrado ser imprecisas y subjetivas. Con la llegada de la era digital se ha producido una auténtica revolución en el campo de la documentación del arte rupestre en general y de los petroglifos en particular. Técnicas como la fotografía digital nocturna o la fotogrametría tridimensional (3D) han abierto un mundo de posibilidades. En el caso de la fotografía nocturna su uso se remonta a la segunda mitad del siglo XX. No obstante, en los últimos años la aparición de cámaras digitales de gran potencia junto con nuevos sistemas de iluminación impensables hace décadas, como las linternas tácticas, han rejuvenecido una técnica que parece incombustible. Tras las numerosas pruebas realizadas por el equipo de trabajo del proyecto DIPAR (Sistema de Documentación Integral del Patrimonio Rupestre) ha quedado de manifiesto que, correctamente empleada, esta técnica sigue teniendo un potencial inmenso en el campo de la documentación de grabados rupestres, especialmente si tomamos en consideracion su bajo coste y facilidad de uso. La fotogrametría por su parte, se muestra como el complemento perfecto a los registros obtenidos mediante la fotografía digital nocturna. La posibilidad de aplicar filtros o shaders que aumentan la visibilidad de los petroglifos digitalizados, dentro de un esquema métrico, supone un gran avance en su documentación. El presente trabajo explora las ventajas asociadas a estas técnicas con un caso práctico: los grabados rupestres inéditos de Alcázar de San Juan (España). Palabras clave: fotografía digital, 3D, patrimonio arqueológico, arte rupestre, fotogrametría, documentación

* Corresponding author: Víctor Manuel López-Menchero Bendicho, [email protected] 64 USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA)

1. Introducción El municipio de Alcázar de San Juan (Fig. 1) se encuadra geográficamente en la cuenca alta del río Guadiana, en pleno centro de La Mancha (Castilla-La Mancha, España). Posee un extenso término municipal, lo que favorece la presencia de un vasto y rico patrimonio arqueológico, legado de las distintas culturas y civilizaciones que a lo largo de la Historia se han asentado en estas tierras (Vaquero et al., 1984). Entre ese patrimonio, algunos investigadores han localizado, en los últimos años, al menos tres estaciones rupestres, a saber: Pozos de Navarro, Pozo del Empego y Pozo de Tello. Todas ellas inéditas y carentes, por tanto, de ningún tipo de documentación o Figura 2: Pozos de Navarro (Alcázar de San Juan). El investigación al respecto. deterioro es constante dado el grado de exposición de los paneles rupestres. En todos los casos se trata de estaciones con grabados rupestres, presumiblemente de época generaciones venideras (Bertilsson, 2003). En este histórica, realizados sobre rocas en disposición contexto, la Universidad de Castilla-La Mancha ha horizontal o ligeramente inclinada. A diferencia de lo trabajado de forma intensiva en la documentación de que ocurre con las pinturas rupestres, localizadas estos grabados. Dichos trabajos se enmarcan dentro generalmente en abrigos, cuevas o covachas, como es de un proyecto de I+D+i titulado “Sistema de el caso del cercano abrigo de La Rendija en Herencia Documentación Integral del Patrimonio Rupestre (Almodóvar, 1994; 2008), estos grabados se (DIPAR)” que, patrocinado por la Junta de encuentran en superficies totalmente expuestas al aire Comunidades de Castilla-La Mancha (POII-2014-005- libre, lo que contribuye a su progresivo deterioro P), ha diseñado, desarrollado y testado una (Fig. 2). De hecho, hoy día, la erosión pero también la metodología integral de registro del arte rupestre, tanto creación de pátinas más o menos homogéneas, junto pintado como grabado, a partir de la aplicación de con diferentes colonias de microorganismos sobre la nuevas tecnologías (Marchante et al., 2016; López- superficie de la roca, impiden poder distinguir a simple Menchero et al., 2016). vista la mayoría de estas manifestaciones rupestres, cuando en origen, suponemos sería relativamente Para el caso concreto de los grabados, la metodología sencillo distinguir estos grabados bajo cualquier de trabajo propuesta se divide en siete fases sucesivas condición lumínica gracias al contraste cromático e interrelacionadas entre sí que abarcan desde la producido por los impactos e incisiones. prospección superficial, tanto diurna como nocturna, hasta el análisis e interpretación final de los grabados Nos encontramos, por tanto, ante un patrimonio (Fig. 3). En el marco de esa metodología, la sensible y frágil que demanda de acciones urgentes de documentación fotográfica nocturna y la documentación integral capaces de asegurar su documentación terrestre 3D forman el nucleo duro de supervivencia, aunque sea en formato digital, para las este sistema de registro.

Figura 1: Mapa de localización de las estaciones rupestres de Alcázar de San Juan (Ciudad Real).

Virtual Archaeology Review, 8(17): 64-74, 2017 65 LÓPEZ-MENCHERO et al., 2017

deterioro que muchas de ellas provocan en los grabados • Prospección superficial (Bednarik, 1979; 2001; Rogerio, 2007; 2009; San Nicolás, 2012, pp. 29-30). Así por ejemplo, la técnica del tizado altera las proporciones de calcio e impide la • Representación cartográfica datación mediante ratio-cationes de los petroglifos. Similar situación parecen provocar los calcos realizados • Documentación fotográfica nocturna con lapiceros o puntas de grafito. Incluso el uso de moldes de látex o la eliminación de líquenes pueden estar provocando alteraciones irreversibles. El problema • Documentación 3D terrestre de base es la introducción de alteraciones químicas o físicas en la roca de consecuencias impredecibles. Es por ello que algunos autores como Loendorf (2001, pp. • Documentación aérea 55-58) apuestan por adoptar posiciones tendentes a la cautela, con objeto de evitar acciones que podrían tener unas consecuencias definitivas. • Sistema de Información Geográfica (SIG) Por consiguiente, resulta prioritario elaborar una metodología de trabajo que minimice esos riesgos al • Análisis e interpretación tiempo que incremente los niveles de objetividad de la información obtenida. Para ello proponemos centrar nuestra atención en lo que algunos han denominado imágenes de segundo orden, es decir “aquellas Figura 3: Metodología diseñada por el proyecto DIPAR para la documentación integral de estaciones rupestres con grabados. inducidas por la imagen fuente en un sensor” (Montero et al., 1998, p. 156; Martín, 2006, p. 134). Entre las múltiples técnicas que se están testando para conseguir 2. La documentación de petroglifos alcanzar ese objetivo (Gutiérrez et al., 2014), en el Hasta la fecha, los sistemas tradicionales de presente artículo nos detendremos únicamente en dos, documentación de arte rupestre han tenido como fin la fotografía digital nocturna y el registro tridimensional último la obtención de calcos de los motivos terrestre, pues consideramos que la combinación de representados, al objeto de facilitar su correcta ambas puede ofrecer unos resultados adecuados a las visualización y consecuentemente su interpretación necesidades de documentación de muchas estaciones (Moneva, 1993). Resulta evidente que la correcta rupestres. visualización de los motivos es esencial para acometer su posterior interpretación con unas mínimas garantías. 3. Fotografía digital nocturna Precisamente por ello, un calco incompleto o impreciso Uno de los mayores problemas a los que debemos puede conducir a interpretaciones erróneas por parte de enfrentarnos en el arduo proceso de documentar su descubridor pero también por parte de aquellos otros grabados rupestres es su falta de visibilidad. Muchos de investigadores que consulten esos mismos calcos en su ellos han sufrido un proceso de erosión tal que resultan búsqueda de paralelos con los nuevos motivos y prácticamente imperceptibles al ojo humano en las estaciones que poco a poco se van estudiando (Tufte, horas centrales del día, tornándose visibles únicamente 1999, pp. 72-73; Bustamante, 2005). En este sentido, al amanecer y especialmente al atardecer, cuando la luz podemos afirmar que en el caso del arte rupestre el incide en ellos de forma rasante (Bustamante, 2005). error en los calcos realizados y publicados, siempre Este fenómeno natural puede ser reproducido de tiene un efecto multiplicador que acrecienta las manera controlada por la noche mediante el uso de luz consecuencias del error cometido y que, a largo plazo, artificial. Desde hace décadas diversos investigadores supone un lastre en la resolución de los múltiples han podido constatar las bondades que el trabajo enigmas que acompañan a este tipo de manifestaciones nocturno ofrece en las tareas de documentación de culturales. A su vez, una documentación detallada y grabados rupestres gracias a la posibilidad de controlar minuciosa, puede ofrecer datos cruciales para la las condiciones lumínicas (Walker et al., 1979; investigación (Ledo, 2016). Como ha puesto de Wainwright, 1985, p. 32; Sanger y Woodward, 1990; manifiesto recientemente Bertilsson (2015), un pequeño Bednarik, 1990; Soto y Rey, 1996; Domingo et al., 2004; matiz puede ser clave para interpretar correctamente un Domingo et al., 2007, pp. 357-359; Villaverde, 2007, pp. grabado, su significado y cronología. 360-361; Burke et al., 2008; Balme y Paterson, 2009, Tradicionalmente, incluso hoy día, se ha apostado por el pp. 63-64; Cortón, 2011, pp. 165-166; Ugalde, 2011, pp. uso de pinturas, tiza u otros componentes para resaltar 33-34; Mas et al., 2015, p. 23). Incluso algunos autores los motivos antes de realizar la toma fotográfica ya que han reivindicando su utilidad frente a técnicas mucho permiten mejorar la visibilidad de los petroglifos más modernas, pero también mucho más costosas y resaltando su interior o su contorno (Royo & Gomez, complicadas como sucede con el escáner laser 3D 1996, pp. 18-20; Seoane-Veiga, 2006, pp. 36-37; (Cassen et al., 2013). No obstante, la metodología más Antczak & Antczak, 2007; Rey & Eiroa, 2009-2010; adecuada de trabajo para la toma de fotografías Falcón, 2013). En otras ocasiones se ha perseguido nocturnas apenas ha sido objeto de discusión por parte conseguir el calco directo del original mediante técnicas de la comunidad científica, lo que ha limitado el alcance invasivas como el frotagge o el calco sobre plástico de esta técnica. Así por ejemplo, el Grupo de Trabajo de (Seoane-Veiga, 2005; Cortón, 2011, pp. 163-165). Las Noruega para la conservación del arte rupestre desventajas e inconvenientes mostrados por todos estos mencionaba su uso en La guía para la documentación, procedimientos ya han sido suficientemente tratados por gestión, presentación y seguimiento del arte rupestre diversos autores, que han centrado sus críticas en la noruego, pero concluía en un tono bastante pesimista falta de objetividad o fidelidad de estas técnicas y en el que “los resultados de la fotografía con luz artificial no

Virtual Archaeology Review, 8(17): 64-74, 2017 66 USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA) siempre son igualmente predecibles”. Dicho informe buen resultado el uso combinado de una apertura de afirmaba también que como regla general, diafragma de f/25 con un tiempo de exposición de 4 s “proporcionan una información más pobre sobre texturas (Fig. 7). Lógicamente estos valores son aproximados y que muestran trazas del uso de herramientas y que dependen del tipo de cámara y objetivo empleados. Sea requieren de una buena iluminación natural” (Bjelland & como fuere, en ambos casos, dado el tiempo de Helberg, 2007, pp. 48-49). Por otro lado, el uso de luz exposición seleccionado, cualquier ligero movimiento de rasante artificial aumenta la subjetividad del motivo la cámara, por muy sutil que sea, provocará que la percibido pues según el ángulo, la potencia y la fotografía salga trepidada por lo que el disparo se debe dirección de la fuente de luz, se realzan unas zonas en efectuar bien con el temporizador de la propia cámara o detrimento de otras, generando a veces formas que bien con un disparador remoto, siendo en cualquier caso pueden resultar engañosas (Moya-Maleno & Hernández, el uso del trípode imprescindible. El hecho de trabajar 2015, p. 1986). Otros autores han señalado algunas de noche, cuando las temperaturas son más bajas, limitaciones en lo que se refiere a la documentación de reduce los niveles de ruido en las fotografías ya que el paneles completos, en lugar de motivos aislados, ya que calor afecta negativamente al sensor de la cámara. Para la necesidad de posicionar la fuente de luz artificial en evitar el efecto de sombras duras y desiguales se ha un solo punto hace prácticamente imposible obtener optado por mover la linterna en órbita parabólica de fotografías de conjunto de todas las figuras que puede 180º alrededor de los paneles y de unos 90º alrededor albergar un mismo panel, pues por lo general la fuente de los motivos, a unos 10-20 cm de altura respecto a la de luz o no puede iluminar el panel completo o no puede superficie del soporte durante el tiempo que dura la iluminar correctamente cada una de las figuras que exposición (Fig. 4). Este procedimiento evita las típicas componen el panel, dada la irregularidad que suele oclusiones en las zonas que quedan totalmente en presentar la superficie de muchas rocas (Seoane-Veiga, sombra en la fotografía nocturna convencional como 2009, p. 37). Pese a estas limitaciones, que deberían consecuencia de la proyección de las propias paredes haber abierto un debate en el seno de la comunidad del grabado o de alguna irregularidad en la superficie de científica internacional, en los últimos tiempos el la roca (Fig. 6). Además permite documentar paneles discurso se ha reducido a mencionar simplemente la completos de más de 1 m de diámetro usando como utilización de esta técnica, sin entrar en detalles. De fuente de iluminación una simple linterna (Fig. 5). hecho, resulta paradójico que se le prestara más atención y páginas a esta técnica en los años ochenta y Una de las ventajas que se han podido registrar en noventa cuando las posibilidades de la fotografía eran campo es que con este sistema solamente hacen falta mucho más limitadas, que a partir de la democratización dos personas para poder ejecutar el trabajo a buen de la fotografía digital. ritmo, siendo posible, incluso, poder llevarlo a cabo un único investigador. La ligereza del equipo, al no utilizar Para suplir todas estas debilidades el equipo de trabajo complicados sistemas de iluminación, permite una gran ha desarrollado una metodología propia cuyos libertad de movimientos lo que redunda en tiempos de resultados consideramos son satisfactorios y ejecución más cortos con menos personal. En el caso reproducibles por otros grupos de investigación. Para que nos ocupa, donde ha sido necesario documentar ello se ha empleado únicamente una cámara réflex más de 40 paneles y 150 motivos distribuidos a lo largo Nikon D90 con 2 objetivos (objetivo NIKKOR 18-105 mm de tres estaciones rupestres distintas, esta clase de VR + objetivo Tamron F004N SP AF 90 mm F/2.8 Di VC aspectos son fundamentales en aras de una mayor USD MACRO 1:1), una linterna LED táctica (con zoom, eficacia y eficiencia en los procesos de trabajo. XM-L de 1600 lúmenes) y un trípode. Tras probar el uso de diversos sistemas de iluminación, como fluorescentes o linternas tradicionales de diversa índole y potencia, la linterna táctica se ha mostrado como el sistema de luz artificial más adecuado dada su potencia, versatilidad, resistencia, durabilidad y precio. Especialmente si tenemos en consideración que durante la toma fotográfica será necesario mover, siguiendo una órbita parabólica, el punto de luz alrededor del motivo o panel que se quiere fotografiar. Para documentar paneles completos es necesario configurar la cámara sin flash, en enfoque manual, con una apertura del diafragma que puede oscilar entre f/22 y f/36 al objeto de lograr una mayor profundidad de campo, una sensibilidad 100 ISO que permite lograr una mayor Figura 4: Al objeto de conseguir una iluminación uniforme el nitidez, y un tiempo de exposición que según el tamaño foco de luz debe moverse describiendo una órbita parabólica de 180º alrededor del panel. El haz dejado por la linterna debe del panel puede oscilar entre los 6 y los 10 s, suficiente quedar fuera del encuadre de la fotografía. En este caso se para que entre una cantidad de luz adecuada y para mantiene el rastro dejado por la linterna al objeto de mostrar la poder mover la fuente de luz artificial alrededor del órbita exacta seguida para documentar este panel. Estación petroglifo. Cuando se trabaja con tiempos de exposición rupestre de Pozos de Navarro. de 10 s es posible reducir la apertura del diafragma hasta valores cercanos al f/36, mientras que cuando el En lo que concierne al registro de las trazas que nos tiempo de exposición es menor, en torno a los 6 s, la pueden ayudar a determinar la técnica empleada para apertura del diafragma debe estar cercana al f/22. La realizar el grabado, consideramos que la luz artificial documentación de los motivos de pequeño y mediano rasante con posición fija es la más apropiada para tamaño permite trabajar con aperturas de diafragma y proceder a su documentación, pues dada su dureza tiempos de exposición menores, habiéndonos dado un resalta cualquier irregularidad de la roca. Para obtener

Virtual Archaeology Review, 8(17): 64-74, 2017 67 LÓPEZ-MENCHERO et al., 2017 un resultado óptimo es imprescindible, por un lado, tiempos de exposición mucho más reducidos, en torno a pegar el punto de luz a la roca de forma que la luz incida los 2 o 3 s. De esta forma ambas técnicas, tanto aquella sobre el grabado lo más rasante posible y, por otro, el que utiliza luz artificial rasante en posición fija como la uso de un objetivo macro que permita recoger todos los que la usa en movimiento, proporcionan resultados detalles resaltados por la iluminación (Fig. 8). complementarios, válidos y útiles en el proceso de Lógicamente en este caso se puede trabajar con documentación de grabados.

Figura 5: La técnica empleada permite iluminar paneles de varios metros de longitud, como el que se muestra en la imagen, de forma rápida y sencilla. Estación rupestre de Pozos de Navarro.

(a) (b) (c) (d) Figura 6: La fotografía a) muestra el motivo iluminado con una luz plana lo que elimina muchos matices de la forma. Las imágenes b) y c) han sido tomadas con luz artificial rasante aumentando la sensación de relieve del grabado pero ocultando algunas zonas del mismo en función de la posición de la luz. La imagen d) es fruto de una iluminación artificial homogénea rasante conseguida gracias al movimiento del foco de luz en el tiempo que dura la exposición de la cámara. Estación rupestre de Pozo del Empego.

Virtual Archaeology Review, 8(17): 64-74, 2017 68 USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA)

(a) (b) Figura 7: La fotografía a) muestra el aspecto del petroglifo con una luz plana, acorde con el nivel de visibilidad diurno que suele presentar. La fotografía b) muestra los resultados obtenidos tras aplicar la metodología de iluminación y registro desarrollada por el equipo de trabajo. Estación rupestre de Pozo del Empego.

por multitud de investigadores en los últimos años, especialmente desde la aparición de programas informáticos como Agisoft PhotoScan que automatizan hasta el extremo el proceso con excelentes resultados (Plets et al., 2012; Plisson & Zotkina, 2015; Williams & Twohig, 2015; Bertilsson, 2015). Para llevar a cabo los trabajos de documentación fotogramétrica de los grabados se ha utilizado una cámara réflex Sigma SD1 Merrill con un objetivo fijo Sigma 50 mm f/1.4 EX DG HSM, montada sobre un trípode Manfrotto de fibra de carbono MT057C4-G. La estabilidad de la cámara está directamente relacionada con la nitidez de las fotografías, por lo que el uso de un trípode de calidad, ligero, estable y con capacidad para superar los 2 m de altura extendido se ha revelado Figura 8: La fotografía tomada con luz artificial rasante de posición fija permite discernir perfectamente dos de las como algo esencial. Durante el proceso de toma de técnicas empleadas en la realización de los petroglifos de Pozo fotografías en campo se ha podido concluir que las del Empego. En la parte de la izquierda de la imagen se puede recomendaciones dadas por Meijer (2015) para apreciar la técnica del piqueteado más abrasión que permite fotografiar petroglifos, resultan eficaces. Desde trabajar obtener líneas claras y continuas. En la zona central y derecha al atardecer cuando la luz incide en ángulo oblicuo de la imagen por el contrario vemos la técnica del piqueteado sobre los grabados, hasta evitar las sombras fuertes disperso que obliga al espectador a unir los puntos pasando por trabajar preferentemente cuando la mentalmente para poder obtener el contorno de la figura. Como superficie de la roca está seca o usar objetivos de focal se aprecia en la fotografía la utilización de esta técnica permite resaltar, no solo los motivos, sino en general toda la textura de fija de 50 mm. En líneas generales, cuanta mayor sea la la roca. Fotografía tomada con objetivo macro 90 mm. calidad de las fotografías (resolución, ruido…) mejor resultará el modelo 3D. 4. El registro 3D A pesar de la utilidad y del potencial de la fotografía digital nocturna, debemos ser conscientes de sus limitaciones como sistema de registro por sí solo, en la medida que todo grabado posee una naturaleza 3D que la fotografía no es capaz de captar. La posibilidad de documentar de forma métrica en tres dimensiones aquello que posee 3 dimensiones es fundamental a la hora de alcanzar unas cotas óptimas de información registrada. La dificultad de acceso que presentan muchas estaciones rupestres, así como la necesidad de implementar sistemas de documentación de bajo coste que permitan documentar a gran escala el inmenso Figura 9: El nivel de visibilidad de muchos petroglifos es número de motivos y paneles que existen en España, prácticamente nulo. La zona señalizada con un recuadro nos induce a apostar por la fotogrametría como técnica esconde la figura de un antropomorfo. Estación rupestre de de documentación 3D prioritaria. Su uso ha sido testado Pozos de Navarro.

Virtual Archaeology Review, 8(17): 64-74, 2017 69 LÓPEZ-MENCHERO et al., 2017

Las fotografías obtenidas han sido procesadas en Agisoft Photoscan Professional Edition (v. 1.2.4) al objeto de obtener el modelo 3D de los paneles y motivos. Para trabajar con toda la información recogida en campo se ha utilizado una workstation Titanium-Mountain, que permite procesar gran cantidad de datos en poco tiempo. Una vez obtenidos los modelos 3D es necesario exportarlos, en nuestro caso en formato OBJ., para posteriormente trabajar sobre ellos en Meshlab. El software gratuito Meshlab incorpora una opción que permite aplicar filtros o shaders que aumentan la visibilidad de los petroglifos digitalizados (Vergne et al., (a) 2011; Vilas-Estevez et al., 2016; Carrero-Pazos et al., 2016). Concretamente se trata del shader de Radiance Scaling, que reproduce, en cierta medida, los resultados que podríamos obtener con la metodología de Reflectance Transformation Imaging (RTI), pero sin el proceso complicado que implica la adquisición RTI (Mudge et al., 2006; Duffy, 2010; Echevarría & Nieves, 2014). La utilización de shaders en Meshlab no reviste ningún tipo de complejidad, pues únicamente es necesario marcar la opción correspondiente (Labertian Radiance Scaling o Lit Sphere Radiance Scaling) y automáticamente el programa muestra los cambios sobre el modelo 3D. Para asegurar una correcta conservación de los datos, (b) todos los archivos del proyecto han sido transferidos a una unidad de almacenamiento Synology DiskStation DS713+. Dada la constante mejora en los programas de fotogrametría, no solamente se ha procedido a preservar el resultado final 3D, sino también todas las fotografías utilizadas para generar los modelos. Las fotografías digitales constituyen la materia prima de nuestro trabajo, siendo una fuente de documentación de primer nivel. Su correcto almacenamiento puede permitir en el futuro emplear este material con nuevos programas mucho más potentes que los actuales, lo que podría permitir alcanzar cotas de detalle inimaginables en la actualidad. Además, de esta forma, en caso de que en el futuro los paneles o los motivos resulten (c) dañados o desaparezcan será posible contar con una Figura 11: a) Resultado de la digitalización mediante valiosa fuente de información. fotogrametría del antropomorfo de los Pozos de Navarro antes de aplicar ningún tipo de shader; b) Resultado obtenido tras la aplicación en Meshlab del Labertian Radiance Scaling; c) Resultado tras la aplicación del shader Lit Sphere Radiance Scaling. Exportar modelo 3D • Tomar fotos en (obj.) • Aplicar shaders campo (Meshlab) Esta metodología de trabajo también ha dado • Generar nube de excelentes resultados en petroglifos realizados con la puntos, malla y texturas (Agisoft técnica del vaciado o relieve, en donde el motivo se PhotoScan) configura a partir de la extracción del material pétreo Archivar en unidad Importar fotos de almacenamiento que queda en el interior del contorno. La Figura 12 muestra el flujo de trabajo realizado para documentar un panel en el que aparece grabado un calvario completo elaborado con técnica de vaciado.

Figura 10: Flujo de trabajo en el registro 3D. 5. Discusión y conclusiones

Para ejemplificar la utilidad de este procedimiento de El arte rupestre es una de las manifestaciones culturales trabajo (Fig. 10), la Figura 11 ilustra los resultados más importantes y extendidas de la Humanidad. En obtenidos en un caso especialmente complejo dado el España, este hecho se reconoce en el Artículo 40.2 de nivel de erosión y la falta de visibilidad del motivo. Como la Ley 16/1985, de 25 de junio, del Patrimonio Histórico se puede apreciar, con esta técnica es posible distinguir Español, que declara Bienes de Interés Cultural todas sin ninguna dificultad la figura humana que está grabada las manifestaciones de arte rupestre existentes en en la roca y que, sin embargo, apenas resulta visible territorio español. Sin embargo, como hemos podido durante el día como muestra la Figura 9. comprobar con las estaciones rupestres de Alcázar de San Juan, no siempre la teoría coincide con la práctica.

Virtual Archaeology Review, 8(17): 64-74, 2017 70 USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA)

resultados mucho más objetivos, a la par que ahorra tiempo de trabajo en comparación con otros sistemas no invasivos de carácter sistemático, como el desarrollado por Bustamante (2005). Las pruebas llevadas a cabo por el equipo de investigación demuestran que la fotografía nocturna sigue siendo una técnica esencial en el proceso de documentación a gran escala de grabados rupestres, a pesar del tiempo trascurrido desde las primeras pruebas efectuadas en esta dirección durante la segunda mitad del siglo XX. Paradójicamente, el número de publicaciones encaminadas a avanzar en la discusión y consolidación de esta metodología es relativamente escaso. La aparente sencillez de la (a) técnica y su extendida utilización a lo largo del tiempo y el espacio por numerosos investigadores han disipado un debate tan importante como necesario. Así pues, queda todavía mucho espacio para la discusión sobre los sistemas de iluminación, los ángulos de trabajo, la configuración de la cámara, las características técnicas de los equipos, etc. Por otro lado, todavía existe margen para el debate en lo que concierne a sus posibilidades de aplicación, pues la implementación de esta técnica no resulta eficaz en todos los casos, especialmente en aquellos grabados que fueron realizados como simples incisiones ultrafinas sobre rocas de carácter cuarcítico. La escasa alteración

de la superficie de la roca provocada por esta clase de (b) grabados explica la falta de efectividad de este sistema y abre la puerta a otras técnicas que pudieran mostrarse más eficaces en la documentación de este tipo de grabados. La fotogrametría, por su parte, sirve de complemento perfecto a las imágenes obtenidas mediante la técnica de fotografía digital nocturna. La utilización de shaders mejora sustancialmente el nivel de visibilidad de los grabados. Además la gratuidad de Meshlab, unido a su facilidad de uso, puede favorecer la implementación masiva de esta metodología en los procesos de documentación de petroglifos. Resulta altamente significativo que dos equipos de investigación diferentes

y sin conexión entre sí, el representado por Vilas- (c) Estevez, Carrero-Pazos y Vázquez-Martínez y el Figura 12: Resultados del proceso de documentación 3D de un representado por los autores que suscriben estas líneas, panel trabajado con la técnica de vaciado: a) Generación del hayamos llegado a la misma conclusión en cuanto a la modelo 3D en Agisoft Photoscan; b) Importación del modelo 3D utilidad de los shaders de Meshlab. Lo que vendría a a Meshlab en formato OBJ; c) Aplicación del shader Labertian corroborar que en verdad supone una línea de trabajo Radiance Scaling. prometedora en el campo de la documentación de petroglifos. Multitud de enclaves con grabados rupestres, especialmente de época histórica, se encuentran No obstante, las posibilidades en Meshlab parecen completamente abandonados y en serio peligro de todavía muy limitadas. En este sentido, la falta de más desaparición. Dada su ubicación en zonas apartadas y investigación básica en este campo impide obtener al aire libre, la mejor manera que tenemos de mejores resultados. Por otro lado, es importante no protegerlos, es documentarlos y darlos a conocer, ya perder de vista que la calidad del modelo 3D es esencial que solo se protege aquello que se valora y solo se para obtener unos resultados óptimos. Los shaders de valora aquello que se conoce. En este sentido, debemos Meshlab mejoran el nivel de visibilidad de los petroglifos ser capaces de desarrollar metodologías de trabajo en los modelos 3D por lo que cuanta mayor sea la útiles y eficaces, implementables a gran escala, y calidad del modelo mejor será el resultado final. No consecuentemente de bajo coste y relativa sencillez. obstante, siempre es deseable encontrar un punto de equilibrio entre la calidad perseguida y el esfuerzo que Los últimos avances tecnológicos nos permiten será necesario realizar para alcanzarla. En esta lógica, desarrollar una amplia gama de técnicas de registro de las pruebas que hemos llevado a cabo nos inducen a grabados rupestres sin necesidad de mantener ningún considerar el objetivo de focal fija de 50 mm como el tipo de contacto físico con ellos, lo que sin duda más idóneo para acometer estos trabajos, pues permite garantiza su mejor conservación a largo plazo. El uso de documentar tanto figuras aisladas como paneles estas técnicas no invasivas no solo resulta respetuoso completos con suficiente nivel de detalle, siempre y con el bien documentado sino que además ofrece unos

Virtual Archaeology Review, 8(17): 64-74, 2017 71 LÓPEZ-MENCHERO et al., 2017 cuando se haga una adquisición adecuada. La permanecen irresolutos en torno a su sentido y utilización de objetivos más potentes, como el macro 90 significado. mm, permite obtener mejores resultados en figuras aisladas, pero resulta poco práctico en la Agradecimientos documentación de paneles completos o superficies de mediano tamaño, dada la pequeña extensión que es La Junta de Comunidades de Castilla-La Mancha y el capaz de cubrir cada fotografía. Fondo Europeo de Desarrollo Regional a través del proyecto «Sistema de documentación integral del Pese a todo, estos avances nos hacen ser optimistas en patrimonio rupestre» (POII-2014-005-P) han hecho cuanto al futuro inmediato del arte rupestre, pues abren posible este trabajo. Deseamos asimismo mostrar la puerta a trabajos de documentación rigurosos y nuestro más sincero agradecimiento a los masivos. Y es que, solamente si somos capaces de investigadores que nos pusieron sobre la pista de estos estudiar con rigurosidad el arte rupestre en su conjunto, enclaves: D. Jesús Lizcano, D. Ángel Vaquero y D. Juan podremos desentrañar los misterios que aún Ángel Ruiz.

Referencias Almodóvar, J. (2008). Arte rupestre en Herencia: La Rendija. Ayto. de Herencia.

Almodóvar, J. (1994). Las pinturas rupestres del abrigo La Rendija de Herencia. In La Edad del Bronce en Castilla-La Mancha: actas del Simposio, 1990, (pp. 315–332). Toledo, España: Diputación provincial de Toledo. Antczak, M. M., & Antczak, A. (eds.) (2007). Los Mensajes confiados a la Roca. Sobre el inventario de petroglifos de la Colonia Tovar de Peter Leitner. Caracas, Venezuela: Editorial Equinoccio. Balme, J., & Paterson, A. (2009). Archaeology in Practice: A Student Guide to Archaeological Analyses. John Wiley & Sons. Bednarik, R. G. (1979). The potential of rock art patination analysis in Australian archaeology. The Artifact, (4), 14–38. Bednarik, R. G. (2001, 1990). Sobre la práctica de tizar y resaltar petroglifos. Rupestre/web, http://rupestreweb.tripod.com/tizado.html Bertilsson, U. (2003). Rock Art at Risk. In Heritage at risk: ICOMOS world report 2002/2003 on monuments and sites in danger. ICOMOS (pp. 236–239). Munich, Alemania: K.G. Saur.

Bertilsson, U. (2015). Examples of application of modern digital techniques and methods Structure from Motion (SfM) and Multi-View Stereo (MVS) for three-dimensional documentation of rock carvings in Tanum creating new opportunities for interpretation and dating. In F. Troletti (Ed.), XXVI Valcamonica Symposium 2015. Capo di Ponte (Bs) Italy September 9 to 12, 2015 (pp. 1–6). Capo di Ponte, Italia: Edizioni del Centro. Bjelland, T., & Helberg, B. H. (Eds.) (2007). A guide to the documentation, management, presentation and monitoring of Norwegian rock art. Norwegian Working Group for Rock Art Conservation. Directorate for Cultural Heritage, Riksantikvaren. Burke, H., Smith, C., & Zimmerman, L. J. (2008). The Archaeologist's Field Handbook: North American Edition. Lanham, USA: Altamira Press. Bustamante Díaz, P. (2005). Relevamiento de sitio arqueológico de Cuz Cuz, IV Región, Chile: Descripción de una experiencia. Parte I. Relevamiento y rescate de los diseños. Rupestreweb, http://rupestreweb.info/bustamante.html Carrero-Pazos, M., Vázquez-Martínez, A., & Vilas-Estevez, B. (2016). AsTrend: Towards a new method for the study of ancient carvings. Journal of Archaeological Science: Reports, 9, 105–119. http://dx.doi.org/10.1016/j.jasrep.2016.06.044. Cassen, S., Lescop, L., Grimaud, V., Querré, G., & Suner, B. (2013). In search of a representation of the Gavrinis megalithic tomb (Morbihan, France). Virtual Archaeology Review, 4(8), 61–63. http://dx.doi.org/10.4995/var.2013.4319 Cortón, N. Carrera, F., De La Peña, A., Neira, A., & Bernaldo De Quirós, F. (2011). Avance al estudio de los petroglifos de Peña Fadiel (Filiel, Lucillo de Somoza, León). Férvedes: Revista de investigación, (7), 105–114. Domingo, R., Guillem, P., Martínez-Valle, R., & Sopena, M. C. (2004). La documentación gráfica: un trabajo en equipo. In P. Utrilla y V. Villaverde (Eds.), Los grabados levantinos del Barranco Hondo (Castellote, Teruel), (pp. 27–36). Zaragoza, España: Gobierno de Aragón. Domingo Sanz, I., Burke, H., & Smith, C. (2007). Manual de campo del arqueólogo. Barcelona, España: Ariel.

Virtual Archaeology Review, 8(17): 64-74, 2017 72 USO COMBINADO DE LA FOTOGRAFÍA DIGITAL NOCTURNA Y DE LA FOTOGRAMETRÍA EN LOS PROCESOS DE DOCUMENTACIÓN DE PETROGLIFOS: EL CASO DE ALCÁZAR DE SAN JUAN (CIUDAD REAL, ESPAÑA)

Duffy, S. M. (2010). Polynomial texture mapping at roughting linn rock art site. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 38-5, 213–217. Echevarría, G., & Nieves, A. (2014). Análisis con RTI (Reflectance Transformation Imaging) en dos sitios con quilcas del Perú: tecnología, deterioro y conservación. Arqueología y Sociedad, (28), 289–304. Falcón, N. (2013). Solar eclipse represented in the petroglyphs of Vigirima (Venezuela). Virtual Archaeology Review, 4(8), 155–158. http://dx.doi.org/10.4995/var.2013.4357 Gutiérrez Alonso, A., Farjas Abadía, M., & Domínguez Gómez, J. A. (2014). Engineering at the service of history: Paleolithic invisible images. Virtual Archaeology Review, 5(11), 14–19. http://dx.doi.org/10.4995/var.2014.4172 Ledo Fernández, F. (2016). Rescate documental de petroglifos y reconstrucción 3D del corredor dolménico de Cubillejo de Lara, Burgos. Virtual Archaeology Review, 7(14), 43–52. http://dx.doi.org/10.4995/var.2016.4522 Loendorf, L. (2001). Rock Art Recording. In D.S. Whitley (Ed.), Handbook of Rock Art Research (pp. 55–79). Walnut Creek, USA: AltaMira Press. López-Menchero, V. M., Marchante, Á., Vincent, M., Cárdenas, Á. J., & Onrubia, J. (2016). Trabajos de documentación digital de grabados rupestres en el término municipal de Alcázar de San Juan (Ciudad Real, España). In J.L. Lerma y M. Cabrelles (Eds.), Proceedings of the ARQUEOLÓGICA 2.0 8th International Congress on Archaeology, Computer Graphics, Cultural Heritage and Innovation, (pp. 528–530). Valencia, España: Universitat Politècnica de València. Marchante, Á., López-Menchero, V. M., Ramiro, R., & Onrubia, J. (2016). Nuevas aportaciones al conocimiento y documentación de los grabados rupestres de Alcázar de San Juan. In F. Alía, J. Anaya, L. Mansilla y J. Sánchez (Eds.), II Congreso Nacional Ciudad Real y su provincia (pp. 62–74). Ciudad Real, España: Instituto de Estudios Manchegos (CSIC). Martín Rodríguez, E. (2006). La aplicación de las nuevas tecnologías al estudio de las manifestaciones rupestres. Revista Tabona, (14), 117–148. Mas, M., Lemjidi, A., Oumouss, A., Torra, G., Maura, R.… & Gavilán, B. (2015). Proyecto Tamanart 2013-2014. Informes y Trabajos, (12), 15–34. Meijer, E. (2015). Structure from Motion as documentation technique for Rock Art. Adoranten, 2015, 66–73.

Moneva, M. D. (1993). Primeros sistemas de reproducción de Arte Rupestre en España. Espacio, tiempo y forma. Serie I, Prehistoria y arqueología, (6), 413–444. Montero Ruiz, I., Rodríguez Alcalde, A. L., Vicent García, J. M., & Cruz Berrocal, M. (1998). Técnicas digitales para la elaboración de calcos de arte rupestre. Trabajos de Prehistoria, 55(1), 155–169. Moya-Maleno, P. R., & Hernández Palomino, D. (2015). Posibles grabados rupestres en el Cerro Castillón (Villanueva de los Infantes, Ciudad Real): iconografía, arqueología y paisaje. ARKEOS (37), 1981–2008. Mudge, M., Malzbender, T., Schroer, C., & Lum, M. (2006). New Reflection Transformation Imaging Methods for Rock Art and Multiple-Viewpoint Display. In The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage (VAST), (pp 195–202). The Eurographics Association. http://dx.doi.org/10.2312/VAST/VAST06/195-202 Plets, G., Verhoeven, G., Cheremisin, D., Plets, R., Bourgeois, J., … & De Reu, J. (2012). The deteriorating preservation of the altai rock art: assessing three-dimensional image-based modelling in rock art research and management. Rock Art Research, 29(2), 139–156.

Plisson, H., & Zotkina, L. V. (2015). From 2D to 3D at macro and microscopic scale in rock art studies. Digital Applications in Archaeology and Cultural Heritage, 2(2-3), 102–119. http://dx.doi.org/10.1016/j.daach.2015.06.002 Rey Castiñeira, J., & Eiroa García, J. J. (2009-2010). Arte de la Edad del Broncelos grabados ruprestres de Cova da Bruxa y Laxe das Rodas (Muros, A Coruña): actuación arqueológica y revisión interpretativa. Anales de prehistoria y arqueología, (25-26), 77–118. Rogerio Candelera, M. A. (2007). Una propuesta no invasiva para la documentación integral del arte rupestre. Sevilla, España: CSIC - Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS).

Rogerio Candelera, M. A. (2009). Análisis de imagen y documentación integral del arte rupestre: una propuesta de futuro. Estudios de Prehistoria y Arqueología en homenaje a Pilar Acosta Martínez (pp. 171–186). Sevilla, España: Universidad de Sevilla. Royo Guillén, J. I., & Gomez Lecumberri, F. (1996). Los grabados rupestres esquemáticos de Los Pozos Bolletes (Montalbán, Teruel). Zaragoza, España: Ayto. de Montalbán.

Virtual Archaeology Review, 8(17): 64-74, 2017 73 LÓPEZ-MENCHERO et al., 2017

San Nicolás, M. (2012). Documentación gráfica del Arte rupestre postpaleolítico para un plan de gestión. In Jornadas técnicas para la gestión del arte rupestre, patrimonio mundial (pp. 23–43). Huesca, España: Comarca de Somontano de Barbastro. Sanger, K., & Woodward, C. (1990). Discovering prehistoric rock art: a recording manual. Calabasas, USA: Wormwood Press. Seoane-Veiga, Y. (2005). Metodología de reproducción de grabados rupestre en Galicia. Levantamiento de calcos sobre plástico. Cuadernos de estudios gallegos, 52(118), 81–115. Seoane-Veiga, Y. (2006). Novos petroglifos no Monte de Rego Novo (Amoedo-Pazos de Borbén). Cuadernos de Estudios Gallegos, 53(119), 31–52. Seoane-Veiga, Y. (2009). Propuesta metodológica para el registro del Arte Rupestre de Galicia. CAPA 23 (Cadernos de Arqueoloxía e Patrimonio). CSIC - Instituto de Ciencias del Patrimonio (Incipit). Soto Barreiro, M. J., & Rey Castiñeira, J. (1996). Una metodología de estudio para petroglifos: resultados en Laxe da Sartaña. Gallaecia, (14-15), 197–221.

Tufte, E. R. (1999). Envisioning Information. Cheshire, Conneticut: Graphics Press. Ugalde Mora, M. F. (2011). Registro de los petroglifos de Catazho (Morona Santiago). In M.F. Ugalde y A. Yépez (Eds.), Investigaciones arqueológicas en Azuay y Morona Santiago (pp. 13–45). Cuenca, Ecuador: INPC. Vaquero, A., Haro, J., Vela, F., Sereno, I., Aguilar, M. J., Sariñena, Y., Salve, M. S., Padilla, M. L., & Ubeda., D. (1984). Apuntes e inventario de arqueología de Alcázar de San Juan y su comarca. Alcázar de San Juan, España: Patronato de Cultura, Ayto. de Alcázar de San Juan. Vergne, R., Pacanowski, R., Barla, P., Granier, X., & Shlick, C. (2011). Radiance scaling for versatile surface enhancement. En D. Aliaga y M.M. Oliveira (Eds.), I3D '10 Proceedings of the 2010 ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, ACM New York, NY, USA (2010), (pp. 143–150). New York, USA: ACM. Vilas-Estevez, B., Vázquez-Martínez, A., & Carrero-Pazos, M. (2016). Going Further: (Re)Discovering rock art carvings with photogrammetric techniques in Galicia (North-West Iberian Peninsula). In A. Ippolito, & M. Cigola (Eds.), Handbook of Research on Emerging Technologies for Digital Preservation and Information Modeling (pp. 175–200). Hershey, PA: IGI Global. http://dx.doi.org/10.4018/978-1-5225-0680-5.ch008 Villaverde, V. (2007). Algunas recomendaciones de Valentín Villaverde para la fotografía del arte prehistórico parietal y mueble. In I. Domingo, H. Burke y C. Smith (Eds.), Manual de campo del arqueólogo (pp. 360–361). Barcelona, España: Ariel. Wainwright, N. M. (1985). Rock art conservation research in Canada. Bolletino del Centro Camuno di Studi Preistorici, 22, 15–46. Walker, D., Hepplewhite, J., & Dickson, G. (1979). Nightlight petroglyph photography: a photographic technique. In D. Lundy (Ed.), CRARA '77: Fourth biennial conference of the Canadian Rock Art Research Associates, October 27-30, 1977, Victoria, British Columbia (pp. 341–364). Victoria: British Columbia Provincial Museum. Williams, K., & Twohig, E. S. (2015). From sketchbook to structure from motion: Recording prehistoric carvings in Ireland. Digital Applications in Archaeology and Cultural Heritage, 2(2-3), 120–131. http://dx.doi.org/10.1016/j.daach.2015.04.002

Virtual Archaeology Review, 8(17): 64-74, 2017 74

Virtual Archaeology Review, 8(17): 75-83, 2017 http://dx.doi.org/10.4995/var.2017.7028 © UPV, SEAV, 2015

Received: December 17, 2016 Accepted: May 20, 2017

ENTORNO 3D PARA EL ANÁLISIS Y LA RECREACIÓN VIRTUAL DE LAS ACTUACIONES ARQUEOLÓGICAS EN CUEVA DE LA COCINA (DOS AGUAS, VALENCIA, ESPAÑA) 3D ENVIRONMENT FOR THE ANALYSIS AND VIRTUAL RECONSTRUCTION OF THE ARCHAEOLOGICAL FIELDWORKS AT COCINA CAVE (DOS AGUAS, VALENCIA, SPAIN) Agustín Diez Castilloa,*, Alfredo Cortell Nicolaub, Oreto García Pucholb, Pilar Escribá Ruizb a Grup de recerca d’Arqueologia del Mediterrani (GRAM), Departament de Prehistòria i Arqueologia, Universitat de València, 46010 Valencia, Spain. [email protected] b Grup de recerca PREMEDOC. [email protected]; [email protected]; [email protected]

Abstract: In this paper we present our procedure for digitising fieldwork information on the fly (data management), and its combination with the virtual reconstruction of the stratigraphy (virtualisation), of the Cueva de la Cocina site in Dos Aguas (Valencia, Spain). The main tool for the Geographic Information System (GIS) implementation has been OpenJUMP, whilst for the three-dimensional (3D) recreation of the cave virtual environment MeshLab, ParaView, CloudCompare and R open software have been used. According to the data recovered during the two last field seasons at the cave -2015 and 2016-, we present the current state of the stratigraphy virtualisation in the excavated sectors. We also provide not only a general view of the cave, but also different points of view to incorporate distinct geomatics tools into archaeological research. The computer treatment of the data collected in the field provides a better understanding of their spatial relations; which in turn facilitates its analysis and interpretation as well as the realisation of virtual profiles. In the same way, the differences in the frequency of materials belonging to adjacent and/or superimposed stratigraphical units, as well as the total quantities, volumetry and density of the artefacts, with respect to their own stratigraphical unit, or even the whole excavated area, can also be analysed. The combination of both approaches -data management and virtualisation- allows us to integrate geographic information technologies in the daily life of the Mesolithic and Neolithic communities, of which the virtual reconstruction of the different test pits carried out in Cueva de la Cocina constitutes a perfect example. Key words: 3D reconstruction, GIS tools, archaeological fieldwork, Mesolithic, Neolithic, Cocina Cave

Resumen: Con este trabajo pretendemos presentar nuestro procedimiento de digitalización de información de campo (gestión de datos) y su imbricación en la reconstrucción estratigráfica virtual (virtualización) de la Cueva de la Cocina (Dos Aguas, Valencia, España). La herramienta principal para la implementación del Sistema de Información Geográfica (SIG) ha sido OpenJUMP, mientras que para la recreación tridimensional (3D) del entorno virtual de la cueva se han utilizado MeshLab, ParaView, CloudCompare y R. De acuerdo con los datos recuperados durante las excavaciones de los últimos años en la cueva -2015 y 2016-, se presenta el estado actual de la virtualización de la estratigrafía en los sectores excavados. Del mismo modo ofrecemos no sólo una visión general de la cueva, sino también diferentes puntos de vista para incorporar distintas herramientas geomáticas en la investigación arqueológica. El tratamiento informático de los datos recogidos en el campo proporciona una mejor comprensión de las relaciones espaciales entre ellos; lo que a su vez facilita su análisis e interpretación así como la realización de cortes virtuales. De la misma manera, se pueden analizar las diferencias en la frecuencia de los materiales que pertenecen a unidades estratigráficas adyacentes o superpuestas, así como las cantidades totales, la volumetría y la densidad de los artefactos, con respecto a su propia unidad estratigráfica u otras. La combinación de ambos enfoques –gestión de datos y virtualización- nos permite integrar las tecnologías de la información geográfica en la vida cotidiana de las comunidades mesolíticas y neolíticas, de las cuales la reconstrucción virtual de los diferentes sondeos realizados en Cueva de la Cocina constituye un ejemplo perfecto. Palabras clave: Entorno 3D, herramientas GIS, excavación arqueológica, Mesolítico, Neolítico, Cueva de la Cocina

1. Introducción estratigráfica de la intervención de campo y su recreación 3D. Para ello se utiliza de forma prominente Presentamos en este trabajo una propuesta el SIG libre OpenJUMP v. 1.9.0 rev.4795 PLUS (The metodológica que integra la información espacial de JUMP Pilot Project, 2015), pero también Qgis v. 2.14.0 una excavación arqueológica en un entorno de trabajo (Qgis Development Team, 2015) y gvSIG Desktop digital con el objetivo de proceder a la reconstrucción v. 2.2 (gvSIG, 2015). Aquel nos permite reconstruir la

* Corresponding author: Agustín Diez Castillo, [email protected] 75 DIEZ CASTILLO et al., 2017 estratigrafía a partir de la toma de datos realizada mediante estación total con más facilidad. Para el mallado 3D se ha utilizado el software libre MeshLab v. 1.3.2 y los paquetes rgl (Adler et al., 2016), alphashape (Lafarge & Pateiro-López, 2016) y VecStatGraphs3D (Felicísimo, Ruiz Cuetos, Polo García, Cuartero, & García Rodríguez, 2016) para R v. 3.2.3 (R Core Team, 2015) y para su visualización ParaView v. 5.0.1-RC1 (Ayachit, 2015), con el fin de proceder a recrear la excavación en un entorno 3D. Uno de los yacimientos que han servido de base de pruebas para la implementación de estas tecnologías es Cueva de la Cocina (Dos Aguas, España). Se trata de un yacimiento arqueológico reconocido en la bibliografía nacional e internacional desde que se realizaran las primeras intervenciones en el año 1941 (García Puchol, Figura 1: Excavación en la Cueva de la Cocina. Juan Cabanilles, McClure, Diez Castillo, & Pardo Gordó, 2015; García Puchol et al., 2014; Fortea, 1973; Pericot, 2. Metodología 1945) dada su amplia secuencia referida principalmente a la prehistoria reciente desde unos niveles basales y La Cueva de la Cocina es probablemente uno de los medios que corresponden al Mesolítico Geométrico (los primeros yacimientos arqueológicos peninsulares en los últimos caza-recolectores en el marco del Mediterráneo que la revolución planteada desde la corriente teórica peninsular) y una secuencia neolítica conservada de conocida como New Archaeology se plasmó en una forma desigual (desde el Neolítico antiguo a la Edad del rigurosidad metodológica que llevó a documentar Bronce). La metodología empleada y la reconstrucción tridimensionalmente todos los hallazgos de la mano del estratigráfica que presentamos se ha implementado profesor Javier Fortea (excavaciones entre 1974 y 1981) durante la primera campaña de actuación arqueológica siguiendo los métodos desarrollados en Francia realizada en el marco del proyecto HAR2012-33111 (Laplace & Méroc, 1954). Gracias a ello, se conserva en “MesoCocina: Los últimos caza-recolectores y el el Servei d’Investigacions Prehistòriques (SIP) de paradigma de la neolitización en el Mediterráneo València el registro manual del inventario de los peninsular” y financiada por la Diputación de Valencia hallazgos por cuadros y capas que incorpora las en el programa de excavaciones 2015 del Museu de coordenadas XYZ de los mismos (referidas la X y la Y al Prehistòria de València. ‘lado derecho’ y el ‘fondo’ de cada uno de los cuadros, mientras que la profundidad se tomaba con un nivel de La intervención arqueológica del año 2015 consistió agua o referida a una estructura metálica fija), así como en la realización de una serie de sondeos inmediatos una descripción genérica de los materiales (fauna, sílex, a las excavaciones realizadas por Pericot en los años cerámica, malacofauna, cantos, plaquetas, ornamento, 40 del pasado siglo (entre 1941 y 1945) y de Fortea entre otros). Todo ello se acompaña de croquis de cada en los años setenta (1974-1981). Previamente, en el una de las plantas sobre papel milimetrado, que refieren año 2014, se llevó a cabo el escaneado 3D de la la posición de los objetos, así como el dibujo de los cavidad a partir de la cual se obtuvieron la planta y las cantos y bloques de piedra de medio y gran tamaño, secciones de la cueva, así como el modelo 3D de la además de fotografías verticales que permiten cavidad con la finalidad de proceder a la correcta comprobar la posición de todos los elementos ubicación de los datos espaciales generados durante señalados. las intervenciones recientes con aquellos disponibles sobre las anteriores actuaciones arqueológicas Las excavaciones actuales han continuado con esta dirigidas por Pericot y Fortea. labor apoyadas ahora en avances tecnológicos que nos permiten sustituir el nivel de agua por la estación total En el trascurso de la campaña de 2015 (Fig.1) fueron en las labores diarias y el escáner Faro 3D para la practicadas un total de 6 catas de desigual dimensión y reconstrucción física del contenedor, llevada a cabo por potencia, mientras que la campaña de 2016 se centró la empresa Global Mediterránea & Geomática. El casi exclusivamente en una única zona. El objetivo escaneado 3D de la cavidad se llevó a cabo mediante prioritario de ambas era la obtención de información una documentación geométrica de alta definición (HDS) estratigráfica, además de proceder a la realización de un que combina técnicas fotogramétricas con técnicas de muestreo sistemático de materiales arqueológicos, escáner láser 3D, una metodología de medida no biológicos y sedimentológicos con el fin de obtener intrusiva que permite capturar información gráfica y información actual para el análisis detallado de la geométrica tanto en 2D como en 3D. El protocolo de compleja secuencia de las ocupaciones en el lugar. La trabajo SIG aplicado, especificado en Diez Castillo, toma de datos de la excavación ha seguido un protocolo Cortell Nicolau y García Puchol (2016), ha permitido preciso que ha consistido en la digitalización de los definir el yacimiento de estudio a través de planos, la datos tridimensionales tanto referidos a las unidades generación de secciones y la modelización de las estratigráficas (UUEE) como a los hallazgos. Este diferentes zonas de la cavidad. protocolo ofrece varias ventajas para la gestión de una excavación arqueológica como la versatilidad y agilidad El modelo digital de superficie (MDS) se ha realizado en la toma de datos y, sobre todo, por lo que se refiere a mediante la utilización de un láser escáner terrestre su precisión, toda vez que abre un amplio abanico de (terrestrial laser scan o TLS, por sus siglas en inglés), nuevas posibilidades para su presentación en diferentes concretamente el modelo FARO FOCUS 3D. El escáner formatos que permiten el análisis eficiente de los datos y láser realiza el barrido de una superficie captando miles su visualización 3D. de puntos por segundo mediante un haz láser en

Virtual Archaeology Review, 8(17): 75-83, 2017 76 ENTORNO 3D PARA EL ANÁLISIS Y RECREACIÓN VIRTUAL DE LAS ACTUACIONES ARQUEOLÓGICAS 2015 EN CUEVA DE LA COCINA (DOS AGUAS, VALENCIA, ESPAÑA)

Figura 2: Reconstrucción de las UUEE 1145 y 1146 con OpenJUMP, incluyendo los puntos de hallazgos tomados en campo. abanico. El instrumento registra toda la nube de puntos 3D compuesta por miles de medidas individuales en un sistema de coordenadas (XYZ) que en sí mismo puede utilizarse para generar un modelo 3D de los objetos registrados. Con una distancia media de escaneado de 10 m, como la aplicadada en nuestro caso, el Figura 3: Modelo de ficha de UE. instrumento puede conseguir una resolución de 2 mm. La imagen resultante es una colección de millones de El primer paso, en todo caso, ha sido la realización de medidas 3D que produce una representación digital una base de datos relacional, compuesta de tres tablas, precisa del entorno. Una vez descargados y guardados en la que por un lado se recoge la descripción de la UE, los datos del escáner, se referencian las nubes de por otro se almacena el material gráfico y, por último, se puntos en un sistema de coordenadas local, creado a almacenan las coordenadas de cada uno de los puntos partir de las cuadrículas dispuestas por Fortea. Todo que se toman en la excavación con la estación total. Las ello se consigue a través de la identificación en la nube relaciones establecidas hacen que la información de puntos de las esferas. A continuación se realiza el geométrica de los puntos se incorpore a la tabla de registro de la nube de puntos, que coloca los UUEE, rellenando campos calculados como la cota escaneados automáticamente en nuestro sistema de mínima o máxima de forma automática, y desde esta se coordenadas y calculará el ajuste con un error de pueda consultar la información gráfica que, correlación de las nubes de puntos de 1.85 mm entre los normalmente, contiene una fotografía de campo de la diferentes escaneados. Finalmente se genera un archivo UE correspondiente y una planta de la misma en la que de intercambio ASCII compatible con aplicaciones de se incluyen los hallazgos (Fig. 3). En la actualidad el entorno de modelización 3D. programa gestor de la base de datos, por razones de Para el tratamiento de la información de las UUEE nos compatibilidad histórica con nuestro grupo de hemos apoyado en varios programas informáticos, todos investigación, es Filemaker (FileMaker Pro® 12), ellos software de código abierto y libres. Los planos de aunque se está preparando una migración de los datos cada una de las UUEE se han realizado con OpenJUMP a PostgreSQL con la extensión espacial PostGIS. A (The JUMP Pilot Project, 2015) dado que el programa partir de la consulta de la tabla de puntos se dibujan las dibuja tridimensionalmente las geometrías a partir de las plantas y se realiza su reconstrucción volumétrica. coordenadas de los puntos tomadas en el campo (Fig. El mallado de las UUEE se ha realizado con MeshLab, 2). OpenJUMP fue uno de los primeros programas SIG programa de manejo rápido y que tiene una amplia en incorporar esa característica, la cual permite la gama de algoritmos para reconstruir volúmenes que interpolación de nodos 3D complementarios si fuera incluyen desde la envolvente mínima hasta la necesario. El dibujo se puede realizar de forma manual decimación de vértices (Fig. 4). uniendo los puntos importados desde la base de datos en la que se vuelcan los datos tomados con la estación Los hallazgos que no se han documentado total o de manera semiautomática gracias a la librería tridimensionalmente en la excavación (los que proceden Sextante (Olaya, 2009). Son varios los algoritmos de de las labores de criba y lavado de sedimentos) se Sextante que permiten la reconstrucción de la superficie generan dentro del volumen de la UE correspondiente de la UE, como la triangulación Delaunay y el algoritmo con una rutina en la que intervienen las librerías foreign de envolvente mínima pero todos ellos conservan la (R Core Team, 2016), spatstat (Baddeley & Turner, referencia altimétrica de los vértices originales dibujando 2005) y alphashape3d (Lafarge & Pateiro-López, 2016) polígonos 3D. del programa R (R Core Team, 2015). A partir de la

Virtual Archaeology Review, 8(17): 75-83, 2017 77 DIEZ CASTILLO et al., 2017

envolvente mínima 3D de los datos tomados en la cueva, se generan de forma aleatoria, tantos puntos como hallazgos, que luego se introducen de manera definitiva en la base de datos. Esta tarea es quizás la más novedosa metodológicamente hablando, porque en un paso previo se genera el volumen de cada una de las UUEE y posteriormente se generan los hallazgos que se han recuperado en cada una de ellas, siendo también posible introducir los volúmenes de las UUEE generados previamente. El proceso no es sencillo porque los volúmenes de las mismas pueden tener formas poliédricas complejas y, por ello, la generación de puntos en su interior no es un proceso simple, hay que comprobar que cada uno de los puntos generados Figura 4: Virtualización estratigráfica 3D preliminar con se halla realmente dentro de la forma 3D de la UE y si MeshLab. es así validarlo hasta completar el número de hallazgos

Figura 5: Resultados del escaneado láser 3D (FARO FOCUS 3D) para la Cueva de la Cocina.

Virtual Archaeology Review, 8(17): 75-83, 2017 78 ENTORNO 3D PARA EL ANÁLISIS Y RECREACIÓN VIRTUAL DE LAS ACTUACIONES ARQUEOLÓGICAS 2015 EN CUEVA DE LA COCINA (DOS AGUAS, VALENCIA, ESPAÑA) necesario, gracias a la función inashape3d del paquete Por otro lado, el tratamiento informático de los datos alphashape3d (Lafarge, Pateiro-López, Possolo, & recogidos en campo proporciona una mejor Dunkers, 2014). Hasta este momento normalmente se comprensión de las relaciones entre los mismos generaban dentro del bounding box 3D (Mousavian, mediante su situación espacial; lo cual a su vez facilita Anguelov, Flynn, & Cosecka, 2017) de cada UE, pero su análisis e interpretación, así como la realización de eso puede generar puntos fuera del volumen de la UE. cortes y capas virtuales, para cuya recreación se ha elegido, en este caso, el software Paraview, al El aspecto metodológico se completa con la considerar la buena relación velocidad-calidad en los consolidación de los hallazgos en la base de datos y un aspectos de visualización, así como sus características protocolo que permite dibujar de forma semiautomática específicamente diseñadas para el ámbito de la las plantas de cada una de las UUEE, varios scripts que investigación (Fig. 7). permiten exportar los datos de la base de datos en formatos que son compatibles tanto con MeshLab, como con OpenJUMP o Qgis.

3. Resultados La obtención del modelo de la cavidad ha propiciado un entorno de trabajo versátil que permite la digitalización de los datos métricos generados en el transcurso del trabajo de campo siguiendo el sistema de coordenadas que en su día planificó el profesor Javier Fortea (Fig. 5). Los sondeos practicados por nosotros han seguido dicho sistema, de forma que será posible conectar la información métrica de los trabajos de campo efectuados en los años 1970, una vez incorporados a las bases de datos diseñadas para tal fin, con los generados en la actualidad. La implementación de este Figura 7: Acabado de las UUEE con Paraview. tipo de tecnologías en el campo arqueológico proporciona interesantes resultados que cabe En la intervención de 2015 en Cueva de la Cocina condensar en un aspecto doble, en el sentido de que practicamos 6 sondeos en los que se definieron un total nos permite una ubicación correcta del yacimiento en de 168 UUEE. Se digitalizaron los datos XYZ de 2192 cuanto a su situación en el espacio, pero también en puntos que definían cada una de las UUEE, así como la referencia a las relaciones que las propias UUEE posición de los materiales que fueron recogidos in-situ guardan entre sí. durante la excavación. Los materiales recuperados en el proceso de flotación de los sedimentos han sido En este sentido, una de las herramientas de provecho incorporados a cada una de las UUEE a través de su es la que nos permite la georreferenciación de distintos generación en el volumen de la UE correspodiente. La ítems que puedan ser de interés para la investigación. digitalización ha supuesto de este modo una mayor Mediante su uso, obtenemos tanto una corrección de la agilidad durante el proceso de toma de datos, así como deformación de proyección inherente a la toma de su proyección digital. De este modo, podemos fotográfica como la geolocalización correcta del observar las diferencias de frecuencia de aparición de elemento en cuestión (Fig. 6). los materiales en UUEE adyacentes o superpuestas (Fig. 8), así como apreciar cantidades totales, volumetría y densidad de los hallazgos localizados, con respecto a su propia UE, o incluso a la totalidad del área excavada. A partir de esta figura podemos visualizar resultados de interés referidos a la distribución del material arqueológico recuperado. La Figura 8 muestra cuatro gráficos 3D en los que queda representada la densidad de diferentes tipos de hallazgos en el sector cuatro de la intervención más reciente. Este sondeo se practicó en un área cercana a la actual entrada de la cavidad sobre una superficie de 4 m2 correspondiente a los cuadros JE 1S, JE 2S, KE 1S, KE 2S. Se excavó en capas sucesivas de 3 cm hasta alcanzar el nivel mesolítico intacto en uno de los cuadros. Los materiales representados se corresponden con los hallazgos de las UUEE de adscripción prehistórica, tanto la unidad CS-3 caracterizada por la presencia de cerámicas neolíticas, como la unidad CS-4 en la que destacan la presencia de microlitos geométricos y la ausencia de material cerámico. Nos interesaba especialmente esta comparación con el fin de explorar el patrón de distribución de los diferentes materiales Figura 6: Reconstrucción fotogramétrica de la UE 1083. entre ambas unidades. El gráfico nos muestra a este respecto resultados interesantes que marcan ciertas

Virtual Archaeology Review, 8(17): 75-83, 2017 79 DIEZ CASTILLO et al., 2017

diferencias destacables. En este sentido, llama la atención ver cómo la distribución de cerámica (Fig. 8a) queda concentrada en un área determinada que no coincide con las áreas de distribución y concentración de fauna (Fig. 8b), malacofauna (Fig. 8c), o de los materiales líticos (Fig. 8d). Este aspecto apunta hacia la diferenciación en las formas de ocupación del espacio y también de acumulación diferencial de los vestigios arqueológicos entre las ocupaciones mesolíticas y neolíticas al menos en esta área de la cavidad. Además de esa distribución divergente del conjunto de cerámicas y de restos líticos, el test de Rayleigh aplicado a cada uno de los conjuntos demuestra, sin embargo, que internamente su distribución es (a) uniforme. Precisamente, esa posibilidad de estudiar una determinada intervención de modo general, o bien de subdividirla en diferentes sondeos que pueden ser relacionados, todo ello en un mismo ámbito de trabajo, es una de las utilidades destacadas en el uso de este tipo de tecnologías. Dada la gran cantidad de información recuperada durante la excavación, cualquier elemento que agilice la gestión de datos supone un gran avance para una interpretación precisa de la información disponible. En cualquier caso, como nos demuestra la ejemplar labor de campo del profesor Javier Fortea en la década de los setenta del siglo pasado registrando más de 8000 piezas arqueológicas in situ, el problema no es sólo (b) registrar la información del material recogido durante la excavación, sino su posterior procesado y es aquí donde un protocolo de actuación como el seguido por nosotros demuestra su potencial (Fig. 9).

4. Perspectivas de futuro Aparte de las grandes posibilidades metodológicas que ofrecen las herramientas SIG para la optimización de la labor de investigación arqueológica que ya son conocidas, su aplicación en otras facetas arqueológicas como la didáctica y la musealización contribuye no solo a una correcta conceptualización del objeto de estudio, sino también a una mayor y mejor difusión de esta disciplina. Uno de los elementos destacados en este sentido es la posibilidad de virtualización 3D, tanto de espacios generales, como (c) de elementos particulares de especial interés. La combinación de ambos enfoques -la gestión de datos y su virtualización- nos permite exprimir e integrar las tecnologías de información geográfica en una paradoja que las funde en el ejemplo presentado con la cotidianidad de las comunidades mesolíticas y neolíticas. La reconstrucción 3D del espacio de Cueva de la Cocina, con la implementación de los sondeos llevados a cabo en la excavación de 2015 constituye un buen ejemplo que sumar a los realizados en los últimos años (Feito Higueruela & Segura Sánchez, 2010; Forte, 2011; Forte, 2014; Tejerina Antón, Bolufer i Marqués, Esquembre Bebia, & Ortega Pérez, 2012; Torres, Cano, Melero, España, & Moreno, 2010). La documentación 3D de yacimientos arqueológicos resulta cada vez más común gracias a las aportaciones tecnológicas de los escáneres 3D que (d) facilitan la documentación 3D de los conjuntos Figura 8: Representación de la densidad 3D de distintos tipos arqueológicos gracias tanto a su “fácil manejo” como a de materiales: a) cerámica, b) malacofauna, c) fauna y d) sílex costos cada vez más asequibles. en el sondeo 4. Realizados con el paquete VectStatGraph3D.

Virtual Archaeology Review, 8(17): 75-83, 2017 80 ENTORNO 3D PARA EL ANÁLISIS Y RECREACIÓN VIRTUAL DE LAS ACTUACIONES ARQUEOLÓGICAS 2015 EN CUEVA DE LA COCINA (DOS AGUAS, VALENCIA, ESPAÑA)

Figura 9: Distribución de los materiales arqueológicos provenientes del sondeo 4 realizada con el paquete rgl.

En la actualidad nuestro equipo de trabajo está 5. Conclusión trabajando en completar la base de datos con la información proveniente de las excavaciones de los La aplicación de técnicas de reconstrucción 3D al profesores Lluis Pericot (1941-1945) y Javier Fortea proceso de excavación de la Cueva de la Cocina ha (1974-1981). El volumen de materiales depositados en demostrado ser un elemento que ayuda tanto en la los fondos del Museu de Prehistòria de Valencia es muy gestión de datos, como en la virtualización de los elevado, dado que estos sondeos se practicaron en el diferentes momentos de ocupación de la cueva. Si bien área de mayor concentración de evidencias de la reconstrucción de yacimientos arqueológicos con ocupación humana en la cavidad. nuevas tecnologías como los escáneres 3D es un hecho cada vez más común, no lo es tanto su aplicación en el Aplicando la misma rutina que a los materiales día a día del proceso de excavación. En este sentido, procedentes del cribado de los sedimentos de 2015, se nuestro protocolo de actuación demuestra que es pueden introducir los materiales no inventariados, posible combinar las técnicas tradicionales de escasos en las excavaciones de Fortea, atendiendo a excavación y documentación con las nuevas tecnologías sus unidades mínimas de excavación, que varía desde en el trabajo arqueológico cotidiano. El trabajo realizado paquetes de 1.5 x 1.25 x 0.30 m (562 l) en los que en Cueva de la Cocina supone, por ello, un pequeño además se mezclan tierras provenientes de distintos paso hacia la normalización del uso de esas tecnologías niveles hasta paquetes de 0.33 x 0.33 x 0.04 m (4 l) a las excavaciones arqueológicas. Las posibilidades que individualizados con criterios estratigráficos. El equipo se abren son inmensas y serán las actuaciones futuras del profesor Javier Fortea inventarió y registró las que pongan de relieve la potencialidad del trabajo de tridimensionalmente más de 8000 piezas resultado de documentación realizado. Los arqueólogos tenemos una sus intervenciones, principalmente referidas al obligación deontológica para con el patrimonio que denominado sector A, cuya inclusión (prácticamente implica, de manera muy especial, la documentación de finalizada) facilitará un análisis pormenorizado de sus los hallazgos realizados durante la excavación. relaciones espaciales y una relectura del libro que ese equipo tan cuidadosamente documentó (Diez Castillo En el caso de la Cueva de la Cocina hemos pasado de et al., 2016). modelos 3D destinados principalmente a las labores de divulgación del patrimonio a incoporar esos datos a las En el aspecto de visualización se va a proceder a la tareas de investigación cotidianas del equipo representación de las capas y hallazgos mediante su arqueológico, eso permite que el esfuerzo de adecuada texturización con el apoyo de programas documentación realizado se pueda plasmar en como MeshLab (Cignoni et al., 2008), CloudCompare diferentes trabajos de investigación. (CloudCompare, 2016) y ParaView (Henderson, 2006). Del mismo modo hemos planificado la migración (o al menos la replicación) de la base de datos desde el Agradecimientos entorno FileMaker al entorno PostgreSQL con su El trabajo presentado se ha llevado a cabo en el marco extensión espacial PostGIS, toda vez que el entorno del proyecto “MESO COCINA: los últimos caza- PostGIS parece el lugar más adecuado para combinar recolectores y el paradigma de la neolitización en el los datos provenientes de nuestros escáneres terrestres mediterráneo occidental” (HAR2012-33111), con los datos tomados con la estación total (Smith & subvencionado por el Ministerio de Economía y Levy, 2012). Competividad, Gobierno de España y el proyecto

Virtual Archaeology Review, 8(17): 75-83, 2017 81 DIEZ CASTILLO et al., 2017

HAR2015-68962, EVOLPAST: Dinámicas evolutivas y Mediterránea & Geomática (Valencia). Los trabajos de patrones de variabilidad cultural de los últimos caza- campo se han llevado a cabo dentro del programa de recolectores y el primer neolítico en el este peninsular investigaciones del SIP-Museu de Prehistòria de la (circa 7000-4500 cal BC). El escaneado láser de la Diputación de Valencia. cavidad ha sido efectuado por la empresa Global

Referencias Adler D., Murdoch, D., Nenadic, O., Urbanek, S., Chen, M., Gebhardt, A., Bolker, B., Csardi, G., Strzelecki, A., & Senger, A. (2016). Rgl: 3D Visualization Using OpenGL. R package version 0.95.1441. https://cran.r-project.org/package=rgl Ayachit, U. (2015). The paraview guide: a parallel visualization application. Nueva York: Kitware, Incorporated. Retrieved from https://www.paraview.org/paraview-downloads/download.php?submit=Download&version=v5.3&type=data&os= all&downloadFile=ParaViewGuide-5.3.0.pdf Baddeley, A., & Turner, R. (2005). Spatstat: An R Package for Analyzing Spatial Point Patterns. Journal of Statistical Software, 12(6). http://doi.org/10.18637/jss.v012.i06

Cignoni, P., Callieri, M., Corsini, M., Dellepiane, M., Ganovelli, F., & Ranzuglia, G. (2008). MeshLab: an open-source mesh processing tool. In V. Scarano, R. De Chiara, & U. Erra (Eds.), Eurographics Italian Chapter Conference (pp. 129-136). The Eurographics Association. http://doi.org/10.2312/LocalChapterEvents/ItalChap/ItalianChapConf2008/ 129-136 CloudCompare Project (2016). CloudCompare (version 2.6) [GPL software]. Retrieved from http://www.cloudcompare.org/ Diez Castillo, A., Cortell Nicolau, A., & García Puchol, O. (2016). Implementación de un SIG en la Cueva de la Cocina (Dos Aguas, Valencia). 10as Jornadas de SIG Libre, en prensa. Retrieved from https://www.academia.edu/26741747/Implementaci%C3%B3n_de_un_SIG_en_la_Cueva_de_la_Cocina_Dos_Agua s_Valencia Diez Castillo, A., García Puchol, O., Basile, M., Cortell Nicolau, A., & Tsanté, N. (2016). Propuesta de análisis tridimensional aplicado a la integración de las diferentes actuaciones arqueológicas realizadas en cueva de la Cocina (Dos Aguas, Valencia). In ARQUEOLÓGICA 2.0-8th International Congress on Archaeology, Computer Graphics, Cultural Heritage and Innovation, (pp. 524-527). Retrieved from http://roderic.uv.es/bitstream/handle/10550/56578/diez_el_al2Proceedings Arqueologica 2.0.pdf Felicísimo, A., Ruiz Cuetos, J. C., Polo García, M. E., Cuartero, A., & García Rodríguez, P. (2016). VecStatGraphs3D: Vector analysis using graphical and analytical methods in 3D. R package version 1.6. Retrieved from https://CRAN.R-project.org/package=VecStatGraphs3D Feito Higueruela, F. R., & Segura Sánchez, R. J. (2010). Herramientas SIG 3D. Virtual Archaeology Review, 1(1), 87–91. http://doi.org/10.4995/var.2010.4795 Forte, M. (2011). Cyber-Archaeology: Notes on the simulation of the past. Virtual Archaeology Review, 2(4), 7–18. http://doi.org/10.4995/var.2011.4543 Forte, M. (2014). 3D archaeology: New perspectives and challenges—the example of Çatalhöyük. Journal of Eastern Mediterranean Archaeology and Heritage Studies, 2, 1–29. http://doi.org/10.13140/2.1.3285.0568 Fortea, J. (1973). Los complejos microlaminares y geométricos del Epipaleolítico mediterráneo español. Salamanca: Universidad de Salamanca (Memorias del Seminario de Prehistoria y Arqueología, 4).

García Puchol, O., Juan Cabanilles, J., McClure, S. B., Diez Castillo, A., & Pardo Gordó, S. (2015). Avance de resultados de los nuevos trabajos arqueológicos en Cueva de la Cocina (Dos Aguas, Valencia): campaña de 2015. Saguntum (PLAV), 47, 251–255. http://doi.org/10.7203/SAGVNTVM.47.7195 García Puchol, O., Juan Cabanilles, J., McClure, S. B., Pascual Benito, J. Ll., Martí Oliver, B., Pérez Ripoll, M., Bernabeu Aubán, J., Pardo Gordó, S., Molina Balaguer, Ll., Carrión Marco, Y., & Diez Castillo, A. (2014). Los últimos caza-recolectores en Cueva de la Cocina (Dos Aguas, Valencia). En R. Sala (Coord.), Los cazadores y recolectores del Pleistoceno y del Holoceno en Iberia y el estrecho de Gibraltar (pp. 370–372). Burgos: Universidad de Burgos/Fundación Atapuerca. gvSIG (2015): Retrieved from http://www.gvsig.com/es/productos/gvsig-desktop Henderson, A. (2006). The ParaView Guide: A Parallel Visualization Application. Nueva York: Kitware, Incorporated.

Virtual Archaeology Review, 8(17): 75-83, 2017 82 ENTORNO 3D PARA EL ANÁLISIS Y RECREACIÓN VIRTUAL DE LAS ACTUACIONES ARQUEOLÓGICAS 2015 EN CUEVA DE LA COCINA (DOS AGUAS, VALENCIA, ESPAÑA)

Lafarge, T., Pateiro-López, B., Possolo, A., & Dunkers, J. P. (2014). R Implementation of a Polyhedral Approximation to a 3D Set of Points Using the α-Shape. Journal of Statistical Software 56(4), 1–19. http://doi.org/10.18637/jss.v056.i04 Lafarge, T., & Pateiro-López, B. (2016). Alphashape3d: Implementation of the 3D Alpha-Shape for the reconstruction of 3D sets from a point cloud. R package version 1.2. Retrieved from https://CRAN.R- project.org/package=alphashape3d Laplace, G., & Méroc, L. (1954). Application des coordonnées cartésiennes à la fouille d'un gisement. Bulletin De La Société Préhistorique Française, 51(1-2), 58–66. http://doi.org/10.3406/bspf.1954.12427 Mousavian, A., Anguelov, D., Flynn, J., & Kosecka, J. (2017). 3D bounding box estimation using deep learning and geometry. CoRR, abs/1612.00496. Retrieved from http://arxiv.org/abs/1612.00496 Olaya, V. (2009). Sextante, la libreria de análisis geoespacial libre: presente y futuro. En III Jornadas de SIG Libre, Girona, España: SIGTE-Universitat de Girona. Retrieved from http://hdl.handle.net/10256.1/1035 Pericot, L. (1945). La Cueva de la Cocina (Dos Aguas). Archivo de Prehistoria Levantina 2(nota preliminar), 39–71. Retrieved from http://www.museuprehistoriavalencia.es/web_mupreva_dedalo/publicaciones/349/es

Qgis Development Team (2015). QGIS Geographic Information System. Open Source Geospatial Foundation Project. Retrieved from http://qgis.osgeo.org R Core Team (2015). R: A language and environment for statistical computing. Retrieved from https://www.R-project.org/ R Core Team (2016). foreign: Read Data Stored by Minitab, S, SAS, SPSS, Stata, Systat, Weka, dBase, .... R package version 0.8-66. Retrieved from https://CRAN.R-project.org/package=foreign Smith, N. G., & Levy, T. E. (2012). Real-time 3D archaeological field recording: ArchField, an open-source GIS system pioneered in southern Jordan. Antiquity, 86(331). http://antiquity.ac.uk/projgall/smith331/ Tejerina Antón, D., Bolufer i Marqués, J., Esquembre Bebia, M. A., & Ortega Pérez, J. R. (2012). Documentación 3D de pinturas rupestres con Photomodeler Scanner: los motivos esquemáticos de la Cueva del Barranc del Migdia (Xàbia, Alicante). Virtual Archaeology Review, 3(6), 64–67. http://doi.org/10.4995/var.2012.4443 The Jump Pilot Project (2015). OpenJUMP SIG – The free and open source Java-based desktop SIG. Retrieved from http://www.openjump.org/

Torres, J. C., Cano, P., Melero, J., España, M., & Moreno, J. (2010). Aplicaciones de la digitalización 3D del patrimonio. Virtual Archaeology Review 1(1), 51–54. http://doi.org/10.4995/var.2010.4768

Virtual Archaeology Review, 8(17): 75-83, 2017 83

Virtual Archaeology Review, 8(17): 84-94, 2017 http://dx.doi.org/ 10.4995/var.2017.5972 © UPV, SEAV, 2015

Received: June 17, 2016 Accepted: February 16, 2017

MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA) 3D MODELLING OF THE SALESIAN MISSION NUESTRA SEÑORA DE LA CANDELARIA CEMETERY (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA) Melisa A. Salernoa,*, Daniel Hereñúa, Romina C. Rigonea, Ricardo A. Guichónb a Instituto Multidisciplinario de Historia y Ciencias Humanas, Consejo Nacional de Investigaciones Científicas y Técnicas, Saavedra 15, 5° piso, C1083ACA Ciudad Autónoma de Buenos Aires, Argentina. [email protected]; [email protected]; [email protected]

b Consejo Nacional de Investigaciones Científicas y Técnicas. Facultad de Ciencias Sociales, Universidad Nacional del Centro de la Provincia de Buenos Aires, Calle 508 N° 881, B7631XAF Quequén, Provincia de Buenos Aires, Argentina. [email protected]

Resumen: La misión salesiana Nuestra Señora de La Candelaria funcionó entre finales del siglo XIX y mediados del siglo XX en Río Grande (Tierra del Fuego, Argentina). Su cementerio aún alberga los restos de indígenas y religiosos que habitaron la institución, y colonos que vivieron en las inmediaciones. Desde hace algunos años, el lugar recibe atención por parte de un proyecto de investigaciones que discute la dinámica de las relaciones coloniales en Tierra del Fuego. El camposanto atestigua una historia profunda de transformaciones, y si bien ha sido declarado parte del patrimonio nacional y provincial, sus condiciones de preservación son relativamente precarias. En este trabajo presentamos la labor realizada para modelar en 3D el cementerio de La Candelaria, combinando información obtenida mediante el levantamiento planialtimétrico y fotográfico del lugar. El resultado constituye un registro detallado y con textura fotorrealista sobre la materialidad del camposanto en el presente. De esta manera, no sólo aporta una herramienta relevante para colaborar en la difusión y protección del patrimonio. También ofrece una plataforma para que en un futuro puedan ser modeladas imágenes históricas del cementerio con diversos propósitos de investigación. Palabras clave: cementerio, modelado 3D, levantamiento planialtimétrico, fotorrealismo, patrimonio, Argentina

Abstract: The Salesian mission Nuestra Señora de La Candelaria operated between the late 19th and the mid-20th centuries in Río Grande (Tierra del Fuego, Argentina). Its cemetery still holds the remains of indigenous and religious people who lived in the institution, and settlers who lived nearby. For some years, the place has received the attention of a research project interested in discussing the dynamics of colonial relationships in Tierra del Fuego. The cemetery has a long history of changes, and even though it was declared part of the national and provincial heritage, its preservation conditions are relatively poor. In this article we present the work carried out to produce a 3D model of the cemetery, combining information obtained by planialtimetric and photographic survey. The result represents a detailed and photorealistic record of the materiality of the cemetery at present. Therefore, it does not only provide a relevant tool to colaborate in the promotion and protection of the archaeological heritage. It also provides a platform for future 3D modelling of historical images of the cemetery, with different research goals. Key words: cemetery, 3D modelling, planialtimetric survey, photorealism, heritage, Argentina

1. Introducción Patagonia (Nicoletti, 2004). El cementerio de la misión (53° 43’ 9.82” S, 67° 47’ 57.07” O) se localiza a 600 m La antigua misión Nuestra Señora de La Candelaria se del complejo central de edificaciones de la institución. localiza a 10 km de la actual ciudad de Río Grande Es un lugar relativamente pequeño, con una superficie (Provincia de Tierra del Fuego, Argentina, extremo sur total de 1548 m2. Sin embargo, atendiendo a los de Sudamérica), en el km 2800 de la Ruta Nacional N° 3 documentos históricos disponibles, allí fueron (53° 43’ 29.01” S, 67° 47’ 56.14” W) (Fig. 1). La enterradas más de 340 personas (Salerno & Guichón, institución se emplazó estratégicamente con el propósito 2016; Salerno, García Laborde, Guichón, Hereñú, & de alcanzar a los selk’nam: un grupo de cazadores- Segura, 2016) (Fig. 2). La mayor parte de las mismas recolectores que tradicionalmente habitaba el interior de fueron selk’nam residentes en La Candelaria. Pero la Isla Grande de Tierra del Fuego (Borrero, 2001). también fueron inhumados miembros de la Entre fines del siglo XIX y mediados del siglo XX, La Congregación Salesiana (salesianos y hermanas de Candelaria formó parte del proyecto “evangelizador” y María Auxiliadora) y otros colonos que llegaron y/o se “civilizador” de la Congregación Salesiana sobre instalaron en las inmediaciones.

*Corresponding author: Melisa A. Salerno, [email protected] 84 MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA)

modelado combina información obtenida mediante el relevamiento planialtimétrico y fotográfico del lugar. Desde hace algún tiempo, las técnicas de modelado 3D vienen ganando fuerza en arqueología (Remondino & Campana, 2014). En Argentina, el uso de las mismas aún se encuentra en una etapa inicial (ver, sin embargo, Acuto & Gifford, 2007; Ávido & Vitores, 2013; Lynch & Corrado, 2014; Vázquez & Días Pais, 2014); y en el contexto regional de Patagonia, su potencial ha sido escasa o nulamente considerado. A pesar de la escasez de antecedentes locales, a lo largo del desarrollo de nuestro trabajo encontramos que el modelado 3D del cementerio podía ser una herramienta útil para responder a diversos desafíos –tanto de orden patrimonial como de investigación. En la actualidad, el camposanto únicamente cuenta con una treintena de tumbas y lápidas aisladas en superficie. El resto de los enterratorios carece de demarcaciones visibles. Si bien ha sido declarado Monumento Histórico Nacional en 1999 (Decreto 64/99), y de interés cultural y arqueológico para la provincia de Tierra del Fuego en 2001, el lugar aún resulta poco conocido para algunos habitantes de Río Grande, y para la mayor parte de los turistas que visitan la misión. Las condiciones de preservación del camposanto son relativamente precarias. Como señalaremos posteriormente, la información obtenida a través de registros históricos y las tareas de campo efectuadas por nuestro propio equipo de investigación permiten comprender que la materialidad del cementerio ha sufrido cambios profundos a lo largo del tiempo –incluso en los últimos años. Estas transformaciones comprenden la pérdida, rotura y relocalización de tumbas y lápidas, con datos significativos sobre la identidad de las personas enterradas y el tratamiento que recibieron tras su muerte (Salerno & Guichón, 2016; Salerno et al., 2016).

Teniendo en cuenta estas circunstancias, el modelado Figura 1: Localización de la misión salesiana Nuestra Señora de La Candelaria y su cementerio (Río Grande, Tierra del 3D del cementerio tiene como objetivo crear un registro Fuego, Argentina). Fuente: Google Earth. detallado y con textura fotorrealista sobre el lugar en el presente. Así esperamos contribuir a: Dirigido por Ricardo Guichón, nuestro proyecto de  En un plano patrimonial: investigación analiza la dinámica de las relaciones coloniales en Tierra del Fuego (Guichón, Casali, García ‒ Favorecer la preservación de información frente a Laborde, Salerno, & Guichón, 2017). Bajo este marco, la la potencial pérdida de evidencia (Díaz, Jiménez, misión Nuestra Señora de La Candelaria ofrece un caso Barreda, Asensi & Hervás, 2015). En el hipotético de estudio interesante. Específicamente, su cementerio caso de que algunas tumbas y lápidas permite explorar la salud de las personas, sus prácticas desaparezcan, resulten relocalizadas o terminen y experiencias en lo que supo ser un contexto de deteriorándose, el modelado propone conservar interacciones cotidianas entre indígenas y occidentales. datos sobre su localización, dimensiones, Hasta el momento, las tareas desarrolladas en torno al morfología, materiales constructivos e inscripciones camposanto incluyeron el relevamiento (levantamiento) a la fecha de registro. Este tipo de información no de su superficie, la exhumación de algunos cuerpos, la sólo resulta relevante a la hora de reconstruir la búsqueda y análisis de fuentes documentales (escritos y historia del cementerio; también es especialmente visuales) sobre los enterratorios y los procesos que sensible para aquellos grupos locales que tienen a pudieron afectarlos (Guichón, Suby, Casali, & Fugassa, sus antecesores inhumados en el lugar (como los 2006; García Laborde, Suby, Guichón, & Casali, 2010; miembros de la colectividad selk’nam, la Salerno & Guichón, 2016). Congregación Salesiana y los descendientes de colonos). En este trabajo presentamos la labor efectuada para modelar en 3D el cementerio de La Candelaria. En ‒ Desarrollar una herramienta potencialmente útil particular, centramos la atención en la superficie del para la difusión y puesta en valor del cementerio camposanto, y en las estructuras y rasgos presentes entre un público amplio (Maqueda & Luque, 2015). sobre la misma (incluyendo el muro perimetral, la puerta Esperamos que –en un futuro próximo– el de acceso, las tumbas y lápidas aisladas, entre otros). modelado permita que personas que conocen poco Por el momento, no consideramos el registro o nada sobre el cementerio, se aproximen al mismo subsuperficial producido durante las exhumaciones. El y tengan la experiencia de recorrerlo sin necesidad

Virtual Archaeology Review, 8(17): 84-94, 2017 85 SALERNO et al., 2017

(a)

(b) Figura 2: Cementerio de La Candelaria: a) vista panorámica; b) imagen del lateral correspondiente con la entrada.

de estar en el lugar. Confiamos en que las imágenes históricas ofrecerá la posibilidad de características fotorrealistas del registro lo tornarán direccionar nuevas exhumaciones consensuadas visualmente atractivo, informando sobre las con los grupos locales. condiciones de deterioro y vulnerabilidad en que se encuentra el lugar en el presente. A partir de ello, Este trabajo se divide en tres secciones. En la primera esperamos que el modelado contribuya a generar contextualizamos el caso de estudio. No sólo ofrecemos consciencia y movilizar voluntades a la hora de información sobre la antigua misión Nuestra Señora de colaborar en la protección de esta porción del La Candelaria, y el lugar que tuvo la muerte en su patrimonio. historia. También describimos los estudios que hasta el momento llevamos a cabo en torno al cementerio. En la  A nivel de investigación: segunda sección presentamos las tareas efectuadas para modelar en 3D el camposanto. De este modo, ‒ Sintetizar en un único registro múltiples datos sobre consideramos el relevamiento de datos en el campo y la materialidad del cementerio, ofreciendo la su posterior procesamiento en el gabinete. Por último, posibilidad de “visitar” cotidianamente el lugar en la tercera sección, planteamos los resultados desde el gabinete (aunque ambos puntos disten a obtenidos y la agenda de trabajo futuro. 2800 km). ‒ Obtener una plataforma potencial para modelar 2. El contexto de estudio imágenes históricas del cementerio. Fotografías y filmaciones tomadas entre 1906 y el presente 2.1. Misión y muerte en Río Grande ofrecen instantáneas del lugar a lo largo del tiempo. Contar con un modelado que permita identificar Si bien Tierra del Fuego fue conocida por los europeos puntos de referencia constantes, medibles y desde el siglo XVI, los contactos con los indígenas de la localizables resulta necesario para procesar dichas región mantuvieron un carácter breve y esporádico imágenes mediante “fotogrametría involuntaria” durante siglos. Tan sólo desde mediados del siglo XIX, (sensu Aparicio, Carmona, Fernández, & Martín, con la fundación de “enclaves” coloniales, las 2014). En un futuro, esta posibilidad permitirá interacciones ganaron intensidad (Martinic, 1973; Belza, evaluar la magnitud de los cambios sufridos por el 1975; Borrero, 2001). Estas circunstancias fueron camposanto. Además, en tanto el lugar ha resultado de la acción combinada de diversos factores: atestiguado la pérdida de tumbas y lápidas, el las políticas expansionistas de los estados nacionales modelado de dichas estructuras y rasgos mediante de Argentina y Chile (que pretendían asegurar su

Virtual Archaeology Review, 8(17): 84-94, 2017 86 MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA) soberanía sobre el territorio), la presencia de capitales murieron 8 salesianos y hermanas de María Auxiliadora. europeos (que se encontraban interesados en la En aproximadamente 90 casos, el lugar del deceso se explotación de recursos auríferos y la cría de ganado produjo fuera de la misión o no pudo ser definido. Sin ovino), y la llegada de misioneros (que buscaban embargo, teniendo en cuenta las fechas de las “civilizar” y “evangelizar” a los indígenas). La labor defunciones, los nombres de los fallecidos y los misional fue llevada adelante por protestantes calificativos asociados a los mismos, es posible que en anglicanos y católicos pertenecientes a la Congregación su mayoría hayan sido colonos. Salesiana. Hacia fines del siglo XIX, la Santa Sede permitió que los 2.2. Investigaciones en el cementerio de La salesianos operaran en Tierra del Fuego bajo el marco Candelaria de una Prefectura Apostólica (Fernández, 2014). La Tal como mencionamos, nuestro proyecto de Casa Central de dicha Prefectura se estableció en Punta investigación discute el impacto de las relaciones Arenas, en 1887. Dos años después, la misión de San coloniales en Tierra del Fuego, poniendo en foco La Rafael fue fundada en Isla Dawson (Chile) con el Candelaria y su cementerio. Los primeros estudios propósito de alcanzar a los alacalufes que habitaban los efectuados se interesaron por la demografía y la canales fueguinos. Bajo el modelo de San Rafael, la epidemiología. Para ello se abordaron documentos misión Nuestra Señora de La Candelaria fue establecida históricos en busca de datos sobre salud, enfermedad, en Río Grande para alcanzar a los selk’nam. En junio de condiciones de habitación, dieta, entre otros (Guichón 1893, los salesianos desembarcaron en Arroyo Gama, et al., 2006; García Laborde et al., 2010; Casali, 2013). donde construyeron un cobertizo para pasar el invierno. Asimismo, se realizaron excavaciones en un área A fines de ese año se trasladaron a Barrancos Negros acotada del cementerio (que no presentaba tumbas ni (Río Grande), donde fundaron un primer establecimiento lápidas en superficie). Contando con los permisos misional. Por encontrarse en terreno anegadizo, la necesarios y el consentimiento de los grupos locales, a institución terminó trasladándose a Los Chorrillos. La lo largo de cinco campañas (desde 2007 a 2013) el segunda instalación funcionó entre 1894 y 1896, cuando proyecto trabajó sobre 33 enterratorios. Los análisis un incendio devoró los edificios centrales. En 1897 se consideraron el registro bioarqueológico humano, e inauguró el establecimiento definitivo en las cercanías incluyeron la determinación de sexo, edad, patologías de Cabo Domingo (Beauvoir, 1915). En este trabajo y grupo poblacional. Sin embargo, a medida que centramos la atención en este último emplazamiento (y transcurrió el tiempo, quedó claro que los grupos cuando referimos a “la misión” hablamos sencillamente locales también estaban interesados en conocer el de él). tratamiento que habían recibido los cuerpos de sus La Candelaria acogió indígenas desde un primer antepasados. momento. Sin embargo, la mayor concentración de población se produjo entre 1899 y 1902 (Casali, 2013). Así terminó desarrollándose una nueva línea de trabajo, En ese entonces, el conflicto entre los estancieros y centrada en las prácticas mortuorias, la dinámica de los indígenas adquirió mayor fuerza. Mientras los identidades, y la memoria y el olvido de los difuntos. El estancieros cercaban las tierras y sus rebaños abordaje de la historia del cementerio se volvió desplazaban a los guanacos, los selk’nam traspasaban fundamental, considerando las transformaciones en su los alambrados y mataban ovejas para alimentarse. La materialidad a lo largo del tiempo (Salerno & Guichón, respuesta de los estancieros incluyó la persecución e 2016; Salerno et al., 2016). El relevamiento superficial incluso el asesinato. En este escenario, La Candelaria del camposanto mostró que la mayor parte de los recibió indígenas en busca de refugio o comida, e enterratorios no se encontraba representada por tumbas incluso algunos otros que fueron trasladados por y/o lápidas visibles en superficie. Asimismo, permitió policías y estancieros. Los misioneros brindaron comprender que las tumbas y/o lápidas existentes instrucción religiosa, pero también enseñaron pertenecían mayoritariamente a occidentales (religiosos castellano, costumbres occidentales y diversos oficios. y otros colonos) que habían fallecido entre 1920 y 1940. Las condiciones de vida en la misión tuvieron un Las tareas de excavación ofrecieron información impacto significativo en la salud, favoreciendo la adicional. Algunos enterratorios presentaron fragmentos dispersión de enfermedades como la tuberculosis enterrados de postes de madera, cuyas secciones (Guichón et al., 2006; García Laborde et al., 2010; superiores pudieron perderse por diversos motivos Casali, 2013). Frente al descenso de la población (descomposición, remoción intencional, etc.). Mientras indígena, La Candelaria dejó de funcionar como tanto, algunos enterratorios mostraron señales de haber misión, y fue finalmente transformada en escuela sido perturbados por otros posteriores, sugiriendo que agrotécnica en 1946. una antigua ausencia de demarcaciones pudo haber impedido identificar la presencia de enterratorios El relevamiento y cruce de diversas líneas de evidencia previos. El análisis de documentos escritos permitió (incluyendo documentos históricos y lápidas presentes dividir la historia del cementerio en dos grandes en el cementerio) permitió obtener información sobre las períodos (Salerno & Guichón, 2016; Salerno et al., personas fallecidas y/o inhumadas en La Candelaria 2016). El primero coincidió con el funcionamiento del (Salerno & Guichón, 2016; Salerno et al., 2016). Para el camposanto (cuando se recibieron activamente período 1897-1947, logramos registrar 343 defunciones. inhumaciones, 1897-1947), y el segundo con momentos Todos los individuos fueron enterrados en la misión, posteriores (1947-presente). Los registros misionales (el exceptuando 10 casos sobre los que tenemos dudas o Libro de Defunciones II, 1896-1947; la Crónica del en los que el cuerpo debió ser dejado en otro Padre Zenone, 1898-1902 y el Diario de la Misión, 1896- emplazamiento. Más de 250 personas murieron en la 1947) resultaron iluminadores sobre el primer período institución. La mayor parte de las mismas habrían sido (Salerno et al., 2016). Entre 1897 y ca. 1920 (etapa indígenas, considerando la población del lugar y las inicial del primer período), predominaron las categorías empleadas para describirlas. Allí también inhumaciones de indígenas, mientras que las de los

Virtual Archaeology Review, 8(17): 84-94, 2017 87 SALERNO et al., 2017 colonos fueron escasas. Durante aquellos años también precisión y potencial para efectuar correcciones frente a se realizaron algunas exhumaciones asistemáticas, nuevos relevamientos. probablemente con el objeto de obtener esqueletos de indígenas para conformar colecciones osteológicas. De En noviembre de 2013 efectuamos un nuevo registro, acuerdo a algunas referencias, para 1920 el cementerio buscando ajustar el trabajo realizado en 2006 y obtener prácticamente no contaba con demarcaciones (nunca datos adicionales para comenzar el modelado del habían existido, se habían perdido o habían sido cementerio. El registro fue realizado durante la removidas) (Gusinde, 1920a, 1920b). Entre ca. 1920 y primavera, ya que si bien la estación es ventosa (con 1947 (etapa final del primer período), creció el número ráfagas que pueden superar los 100 km/h), durante el de enterratorios de colonos, a la vez que disminuyó el otoño/invierno las bajas temperaturas y las de indígenas. A partir de ese entonces, los cronistas precipitaciones en forma de nieve impiden el trabajo de comenzaron a efectuar repetidas menciones a la campo. Por un lado, chequeamos los datos colocación de lápidas y tumbas. planialtimétricos obtenidos durante 2006, identificando cambios en el cementerio (como la remoción y Fotografías y filmaciones históricas ofrecen mayoritaria relocalización de tumbas y lápidas). Por otra parte, –aunque no exclusivamente– información sobre el efectuamos un relevamiento fotográfico detallado, tanto segundo período. La caída en las inhumaciones se del terreno como de las estructuras y rasgos visibles en vinculó con la creación de un nuevo cementerio. Entre superficie. Esto resultó relevante para obtener fines de la década de 1960 y principios de 1980, el mediciones adicionales y poder texturizar el modelado camposanto experimentó una remodelación intensiva. de manera fotorrealista. De este modo, se reemplazó el cerco perimetral de madera por uno de cemento; se quitaron las tumbas de Las fotografías y las filmaciones fueron tomadas madera y se mantuvieron las de metal, cemento y mediante criterios pautados. Por un lado, registramos mármol; se distribuyeron cruces blancas de madera, sin los cuatro laterales del camposanto, situando la cámara que necesariamente coincidieran con las tumbas y en el centro del predio. Por otra parte, registramos cada lápidas removidas. Desde principios de la década de una de las estructuras y rasgos en superficie, colocando 1980, cuando Ricardo Guichón efectuó sus primeras la cámara a una distancia de 2 m (en unos pocos casos, visitas a La Candelaria, se produjeron otros cambios, existieron variaciones como resultado de las como la pérdida, relocalización y rotura de algunas dimensiones y la localización de las tumbas y lápidas). tumbas y lápidas. Ello fue resultado de diversos Para recorrer el lugar, utilizamos la estrategia de factores: el paso del tiempo y la falta de mantenimiento, cuadrículas previamente definida en 2006. Las el ingreso de caballos que habitaban en las fotografías y las filmaciones fueron obtenidas por un inmediaciones del predio, etc. mismo observador, en un mismo día de condiciones estables (parcialmente despejado). 3. Modelado 3D del cementerio En el caso de las fotografías, utilizamos una cámara Panasonic Lumix FZ-70, de 5 megapíxeles con un La creación de un modelado 3D detallado y con textura sensor CCD de 1/2.5”. Las imágenes fueron obtenidas fotorrealista del cementerio de La Candelaria dependió con una distancia focal mínima de 3.58 mm, y la de dos instancias: la obtención de datos en el campo, y resolución fue configurada en modo alto. Para cada el procesamiento digital de los mismos en el gabinete. lateral del cementerio tomamos secuencias de 6 Para obtener información sobre las elevaciones y imágenes, teniendo en cuenta como puntos de depresiones del terreno, así como sobre la localización, referencia las diversas secciones del muro perimetral. dimensiones, morfología y aspecto general de las Asimismo, para cada una de las estructuras y los rasgos estructuras y rasgos, recurrimos al relevamiento en superficie, obtuvimos dos conjuntos de imágenes planialtimétrico y fotográfico. Mientras tanto, para (Andrews et al., 2010). En primer término, fotografiamos procesar digitalmente los datos efectuamos tareas de cada una de las caras de los objetos, situando la modelado, texturizado y renderizado (Flores, Romero, cámara a 90° de las mismas. Teniendo en cuenta las Martínez, & Fernández, 2013; Talaverano, 2014). características de los rasgos (tamaño, condiciones de Si bien en la actualidad se encuentran disponibles preservación, tipo de tumbas), no siempre pudimos diversas herramientas para modelar sitios arqueológicos obtener imágenes axiales. En segundo lugar, tomamos (incluyendo el uso de escáner láser), nuestro proyecto fotografías que permitieron captar cada uno de los no cuenta con las mismas. Las actividades vértices de los objetos, situando la cámara a 45°. contempladas demandaron tiempo y paciencia para En el caso de la filmación utilizamos una cámara Sony relevar y procesar los datos. Sin embargo, HDR-SR12E, de 5 megapíxeles con un sensor CMOS representaron una opción de bajo costo, capaz de 1/3.1” ClearVid. Las imágenes fueron tomadas con una responder de manera adecuada a los objetivos distancia focal mínima de 4.9 mm, y la resolución fue propuestos (incluyendo el texturizado fotorrealista). configurada en alta definición. El registro fue considerado complementario del fotográfico. Así 3.1. Obtención de datos buscamos obtener una serie de fotogramas que pudieran encontrarse disponibles en caso de requerir En 2006, los miembros del equipo de investigación imágenes adicionales. Además de obtener capturas de efectuaron un primer relevamiento superficial del los laterales del cementerio, filmamos cada una de las camposanto (García Laborde et al., 2010). Para efectuar estructuras y rasgos en superficie (rodeándolos para un trabajo sistemático, el predio fue dividido en 16 captar sus caras y vértices). cuadrículas. El terreno, las estructuras y los rasgos fueron registrados planialtimétricamente. Los datos Como síntesis de los datos obtenidos en el campo, obtenidos fueron volcados en AutoCAD, no sólo por la podemos señalar que el cementerio de La Candelaria facilidad que el programa ofrece para procesar se localiza a unos 6 m.s.n.m., sobre una planicie de coordenadas espaciales, sino también por su nivel de cordones de grava y arena que representa un

Virtual Archaeology Review, 8(17): 84-94, 2017 88 MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA) ambiente litoral regresivo (Montes, Salemme, un modelado básico, y teniendo en cuenta la Santiago, & Guichón, 2012). El lugar dibuja un información provista por las fotografías, comenzamos a rectángulo de 36 x 43 m. Dos de sus laterales esculpir las superficies de tumbas y lápidas, reproducir corresponden con la entrada y el fondo del predio; las grietas y roturas en los bloques de cemento y mientras que los dos restantes se alinean con la Ruta mármol, modelar las irregularidades en las rejas de Nacional N°3 y el mar. La diferencia entre las cotas metal, etc. Aquí se volvieron importantes diversos más altas y las más bajas del terreno no suele superar complementos de SketchUp, como Solid Inspector, los 50 cm. En 2013, se registraron 32 tumbas1 (de Weld, Sculpt Tools, Subdivide and Smooth, Shape mármol, perimetrales cubiertas y monolito; de Bender, Bezier Spline Curves, Curviloft, Curvizard, cemento, perimetrales cubiertas, descubiertas y Bool Tools. monolito; de rejas de metal, perimetrales), y tres lápidas aisladas2 (de mármol). También se documentaron una gran cruz (de metal, relativamente cercana al centro del predio, con una placa que menciona a la “raza indígena” que habitó Tierra del Fuego), un muro perimetral (compuesto por secciones de cemento y barrotes de madera), una puerta de entrada (con doble tranquera de metal, con un arco rematado por una cruz con un Cristo), un cartel de ingreso (de metal, con una referencia a los “pioneros” de Río Grande), entre otros (Fig. 3).

3.2. Procesamiento de datos Para efectuar el modelado 3D del camposanto (a) empleamos una Apple MacBook 15.4”, con Intel Core i7, 8GB RAM, Nvidia GeForce GT 650M, 1024 Mb; y un mouse 3DConnexion SpacePilot Pro. Los programas utilizados incluyeron FileMaker Pro 14, AutopanoGiga 4.2, SketchUp Make 2015, Photoshop CS4, Blender 2.76, 3ds Max 2017, GraphicConverter 5.9.5, DxO ViewPoint 2 y DxO Perspective 1. Para obtener mejores resultados, recurrimos al empleo de diversos complementos, tanto en SketchUp como en Photoshop. En cuanto al relevamiento planialtimétrico, comparamos los datos obtenidos en 2006 y 2013, efectuando los ajustes necesarios en AutoCAD 2017 (Fig. 4). En cuanto (b) al relevamiento fotográfico, obtuvimos un total de 134 fotografías y seis videos con una duración de 31:06 min. Los archivos fueron ordenados en una base de datos confeccionada con FileMaker. Teniendo en cuenta las imágenes tomadas para cada uno de los laterales, elaboramos una fotografía panorámica del predio con AutopanoGiga. Junto con el documento de AutoCAD, la misma resultó de utilidad para orientarnos de manera rápida y sencilla. El modelado del terreno, las estructuras y rasgos demandó un trabajo intensivo (Fig. 5). Las elevaciones y depresiones, y los volúmenes de las tumbas y (c) lápidas fueron dimensionados a partir de datos planialtimétricos. A partir de ello, definimos puntos y segmentos con el propósito de construir una malla poligonal (Roskes, 2009; Brixius, 2010; Schreyer, 2012; Tal, 2009, 2013). A cada uno de los rasgos añadimos su geolocalización como metadato, de forma de poder emplearlos en proyectos de realidad aumentada. Una vez alcanzado

1 Entendemos las tumbas como las estructuras que se colocan sobre el suelo con el propósito de coronar un enterratorio (pudiendo estar o no acompañadas de lápidas). Vale la pena (d) mencionar que, bajo dos de las tumbas registradas en el Figura 3: Diversos tipos de rasgos registrados en el cementerio fueron enterradas dos personas en lugar de una sola. cementerio: a) tumbas de cemento; b) mármol; c) rejas; 2 Las lápidas aisladas son placas que no pueden ser d) lápida de mármol. claramente asociadas a un enterratorio.

Virtual Archaeology Review, 8(17): 84-94, 2017 89 SALERNO et al., 2017

Figura 4: Planta del cementerio en AutoCAD. Efectuada a partir del relevamiento de 2006, y conteniendo los ajustes del registro de 2013.

Para el acabado fotorrealista de las estructuras y rasgos En breve, esperamos presentar el modelado del tuvimos en cuenta las imágenes obtenidas en el cementerio a los grupos directamente comprometidos relevamiento fotográfico. De esta manera, procedimos a con la historia del lugar (autoridades locales, tomar las texturas de cada una de sus caras. Las provinciales y nacionales; miembros de la fotografías fueron rectificadas (Pérez, Mozas, Cardenal, Congregación Salesiana; representantes de pueblos & López, 2011). Para esta tarea recurrimos al empleo de originarios y descendientes de colonos). Si estos DxO ViewPoint. sectores acceden, el modelado 3D podrá ser subido a la web para garantizar su libre acceso y difusión. Este Finalmente, procedimos al renderizado del modelado tipo de experiencias permitirá evaluar el potencial del (Fig. 6). modelado a la hora de contribuir en la difusión y puesta en valor del patrimonio. Actualmente, hemos 4. Resultados y agenda a futuro preparado una versión más liviana del registro para agilizar su manejo por parte del público. Además, A partir de los relevamientos efectuados en el campo y hemos alcanzado la visualización del modelo mediante el procesamiento digital de los datos, consideramos realidad aumentada. haber alcanzado el objetivo de crear un modelado 3D detallado y con textura fotorrealista del cementerio de La Recientemente, hemos comenzado a trabajar en la Candelaria. De esta manera, el resultado obtenido reconstrucción del cementerio para otros momentos, constituye un registro exhaustivo de la materialidad del recurriendo al empleo de fotografías y filmaciones camposanto en el presente. históricas. Como primer paso, decidimos considerar un segmento del film documental Los Onas; Vida y Muerte Por un lado, el modelado sintetiza información de en Tierra del Fuego (Chapman & Montes, 1977), cuyo calidad sobre la localización, dimensiones y morfología rodaje se llevó a cabo en 1968. Este material resulta de del terreno, y las estructuras y rasgos en superficie. Por interés por diversos motivos: 1) ofrece capturas de otra parte, la textura fotorrealista ofrece información diferentes sectores del cementerio desde diversos sobre los materiales con que fueron construidas las ángulos; 2) aporta información sobre momentos previos tumbas y las lápidas, su estado de conservación, a la remodelación integral del camposanto (antes que se inscripciones, entre otros. Así desarrollamos una removiera una importante cantidad de tumbas y lápidas herramienta para poder aproximarnos a las metas de entre finales de la década de 1960 y principios de 1970 orden patrimonial e investigación que persigue el –ver “Investigaciones en el cementerio de La proyecto. Candelaria”).

Virtual Archaeology Review, 8(17): 84-94, 2017 90 MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA)

(a)

(b) Figura 5: Modelado de rasgos del cementerio (fotografía, modelo inicial y renderizado): a) tumba de rejas; b) conjunto de tumbas de cemento y marmol.

El modelado del camposanto representado por el film de Entre otras cosas, esperamos conocer si en el área Chapman & Montes (1977) tomará como punto de excavada por el proyecto (donde, al momento de partida el registro del lugar en el presente. Atendiendo a efectuadas las tareas de campo, no se registraron puntos de referencia constantes, procurará localizar y tumbas o lápidas en superficie) existieron o no dimensionar las estructuras y rasgos actualmente demarcaciones. Como los individuos recuperados en el ausentes. Esto no sólo permitirá obtener información sector corresponden mayoritariamente con indígenas, y sobre el posible número, diversidad y distribución de las tumbas y lápidas actualmente presentes en el tumbas para 1968. Su comparación con el registro del cementerio corresponden con religiosos y colonos, sería cementerio en el presente permitirá identificar cambios y interesante conocer si –al menos en los casos continuidades en el lugar. evaluados– existieron diferencias en las formas de

Virtual Archaeology Review, 8(17): 84-94, 2017 91 SALERNO et al., 2017

Figura 6: Renderizado de dos vistas de acuerdo al posicionamiento de dos sujetos hipotéticos en el cementerio. demarcar los cuerpos. Finalmente, el mayor o menor continuar estudiándolos una vez que los restos hayan ajuste entre la localización de los enterratorios sido restituidos a la provincia. excavados y aquella de los rasgos que pudieron coronarlos permitirá definir el potencial de las En síntesis, el modelado 3D del cementerio de La reconstrucciones para orientar nuevas exhumaciones Candelaria surgió como respuesta a diversos desafíos, consensuadas con los grupos locales. abriendo simultáneamente numerosas posibilidades de trabajo. Los avances presentados por este artículo En un futuro próximo, la reconstrucción virtual de la representan un primer paso, y esperan poder ser superficie del cementerio podrá ser complementada con profundizados en un futuro inmediato. el modelado 3D de los contextos excavados. Si bien el registro de las campañas no se efectuó específicamente Agradecimientos con este propósito, confiamos en poder efectuar una reconstrucción parcial a través de las imágenes A los miembros del equipo de investigación; disponibles (ver, por ejemplo, la propuesta de Ávido & especialmente, a Martín Fugassa, Pamela García Vitores, 2013; Aparicio et al., 2014). En este momento, Laborde y Patricia Palacios por facilitarnos información las técnicas de modelado 3D están siendo utilizadas en sobre el primer relevamiento del cementerio. A los el marco del proyecto con el propósito de documentar colectivos selk’nam, al “Centro de Antiguos Pobladores”, esqueletos y objetos recuperados en las excavaciones. a la escuela salesiana “Nuestra Señora de la Las técnicas empleadas ofrecerán la posibilidad de Candelaria”, a las autoridades municipales y

Virtual Archaeology Review, 8(17): 84-94, 2017 92 MODELADO 3D DEL CEMENTERIO DE LA MISIÓN SALESIANA NUESTRA SEÑORA DE LA CANDELARIA (RÍO GRANDE, TIERRA DEL FUEGO, ARGENTINA) provinciales. Al Consejo Nacional de Investigaciones Alejandro Ferrari (CONICET-Universidad Austral) por Científicas y Técnicas (CONICET). brindarnos una ayuda desinteresada para mejorar el A los evaluadores anónimos y al editor de Virtual modelado, y construir puentes que faciliten el diálogo Archaeology Review, cuyos comentarios ayudaron a entre los arqueólogos y los técnicos en informática. Las repensar sustancialmente la forma de contar nuestra ideas presentadas en este artículo son de nuestra experiencia de trabajo. Agradecemos especialmente a exclusiva responsabilidad. Referencias Acuto, F. A., & Gifford, C. (2007). Lugar, arquitectura y narrativas de poder: Experiencia y percepción en los centros Inkas de los Andes del Sur. Arqueología Suramericana, 3(2), 135–161. Andrews, D., Bedford, J., Blake, B., Bryan, P., Cromwell, T., & Lea, R. (2010). Measured and drawn: techniques and practice for the metric survey of historic buildings. Swindon: English Heritage. Retrieved from http://archaeologydataservice.ac.uk/archiveDS/archiveDownload?t=arch-1416-1/dissemination/pdf/9781848021709.pdf Aparicio, P., Carmona, J., Fernández, M., & Martín, P. (2014). “Fotogrametría involuntaria”: rescatando información geométrica en 3D de fotografías de archivo. Virtual Archaeology Review, 5(10), 11–20. http://dx.doi.org/10.4995/var.2014.4205 Ávido, D., & Vitores, M. (2013). El archivo fotográfico como fuente para la reconstrucción tridimensional. In V Congreso Nacional de Arqueometría. Rosario, Argentina. Retrieved from http://goo.gl/3PrFX0 Beauvoir, J. (1915). Los Shelknam. Buenos Aires: Librería del Colegio Pío IX. Belza, J. (1975). En la Isla del Fuego. Colonización. Buenos Aires: Instituto de Investigaciones Históricas de Tierra del Fuego. Borrero, L. (2001). Los Selk’nam. Buenos Aires: Galerna. Brixius, L. (Ed.) (2010). Google SketchUp Workshop: Modeling, Visualizing, and Illustrating. Burlington: Focal Press. Casali, R. (2013). Conquistando el Fin del Mundo. La Misión La Candelaria y la Salud de la Población Selk’nam (Tierra del Fuego, 1895–1931). Rosario: Prohistoria. Chapman, A., & Montes, A. (Directors). (1977). Los Onas. Vida y Muerte en Tierra del Fuego [Documentary film]. Argentina: Comité Argentino del Film Antropológico. Crónica del Padre Zenone (1898–1902) (Unpublished documentary record). Available in Museo Histórico y Natural Monseñor Fagnano (La Candelaria, Río Grande, Tierra del Fuego). Decreto 64/99. Monumento y lugares históricos. Declárense a edificios y sitios históricos de la Provincia de Tierra del Fuego, Antártida e Islas del Atlántico Sur. Retrieved from http://servicios.infoleg.gob.ar/infolegInternet/anexos/55000- 59999/55843/norma.htm Diario de la Misión (1896-1947) (Unpublished documentary record). Available in Museo Histórico y Natural Monseñor Fagnano (La Candelaria, Río Grande, Tierra del Fuego). Díaz, F., Jiménez, J., Barreda, A., Asensi, B., & Hervás, J. (2015). Modelado 3D para la generación de patrimonio virtual. Virtual Archaeology Review, 6(12), 29–37. http://dx.doi.org/10.4995/var.2015.4150 Fernández, A. (2014). Con Letra de Mujer. La Crónica de las Hermanas de María Auxiliadora en la Misión Nuestra Señora de la Candelaria (Tierra del Fuego – Argentina). Buenos Aires: EDBA.

Flores Gutiérrez, M., Romero Soria, A., Martínez García, J., Ramallos Asensio, S., & Fernández Díaz, A. (2013). Visualización virtual de La Quintilla. Virtual Archaeology Review, 4(9), 143–147. http://dx.doi.org/10.4995/var.2013.4265 García Laborde, P., Suby, J., Guichón, R., & Casali, R. (2010). El antiguo cementerio de la misión de Río Grande, Tierra del Fuego. Primeros resultados sobre patologías nutricionales-metabólicas e infecciosas. Revista Argentina de Antropología Biológica, 12(1), 57–69. Guichón, R., Suby, J., Casali, R., & Fugassa, M. (2006). Health at the time of Native-European contact in Southern Patagonia. First steps, results and prospects. Memoria Instituto Oswaldo Cruz, 101(2), 97–105. Guichón, R., Casali, R., Laborde, P. G., Salerno, M., & Guichón, R. (2017). Double coloniality in Tierra del Fuego, Argentina: A bioarchaeological and historiographical approach to Selk’nam demographics and health (La Candelaria mission, late 19th and early 20th centuries). In M. Murphy & H. Klauss (Eds.), Colonized Bodies, Worlds

Virtual Archaeology Review, 8(17): 84-94, 2017 93 SALERNO et al., 2017

Transformed: Toward a Global Bioarchaeology of Contact and Colonialism (pp. 197-225). Gainesville: University Press of Florida. Gusinde, M. (1920a). Expedición a la Tierra del Fuego. Publicaciones del Museo de Etnología y Antropología de Chile, 2(1), 9–44. Gusinde, M. (1920b). 2do Viaje a la Tierra del Fuego. Publicaciones del Museo de Etnología y Antropología de Chile, 2(2), 133–164. Libro de Defunciones II (1896–1947) (Unpublished documentary record). Available in Museo Histórico y Natural Monseñor Fagnano (La Candelaria, Río Grande, Tierra del Fuego). Lynch, J. & Corrado, G. (2014). Arqueología virtual aplicada al sitio Villavil, Catamarca, Argentina. Virtual Archaeology Review, 5(11), 86–92. http://dx.doi.org/10.4995/var.2014.4184 Martinic, M. (1973). Panorama de la colonización en Tierra del Fuego entre 1881-1900. Anales del Instituto de la Patagonia, 4, 5–69. Maqueda García-Morales, R., & Luque Cortina, M. (2015). Paleocatálogo 3D: Fotogrametría para la realización de un catálogo virtual en 3D de alta calidad, accesible y gratuito. Virtual Archaeology Review, 6(13): 35–40. http://dx.doi.org/10.4995/var.2015.4369 Montes, A., Salemme, M., Santiago, F., & Guichón, R. (2012). Prospección con GPR (ground penetrating radar) en el cementerio de la misión salesiana, Río Grande (Tierra del Fuego, Argentina). Comechingonia, 16, 261–272. Retrieved from http://www.comechingonia.com/COMECHINGONIA%20IMPRESA/Montes%20et%20al%202012.pdf Nicoletti, M. (2004). La Congregación Salesiana en la Patagonia: “Civilizar”, educar y evangelizar a los indígenas (1880- 1934). Estudios Interdisciplinarios de América Latina y el Caribe 15(2). Retrieved from http://eial.tau.ac.il/index.php/eial/article/view/894/943 Pérez García, J., Mozas Calveche, A., Cardenal Escarcena, F., & López Arenas, A. (2011). Fotogrametría de bajo coste para la modelización de edificios históricos. Virtual Archaeology Review, 2(3), 121–125. http://dx.doi.org/10.4995/var.2011.4633 Remondino, F. & Campana, S (Eds.). (2014). 3D Recording and Modelling in Archaeology and Cultural Heritage. Theories and best Practices. Oxford: Archaeopress. Roskes, B. (2009). Google Sketchup Cookbook: Practical Recipes and Essential Techniques. Sebastopol: O'Reilly Media. Salerno, M., & Guichón, R. (2016). Sobre la memoria y el olvido: Los difuntos selk’nam y el cementerio de la misión salesiana Nuestra Señora de La Candelaria (Río Grande, Tierra del Fuego). Magallania. In press. Salerno, M., García Laborde, P., Guichón, R., Hereñú, D., & Segura, M. (2016). Prácticas mortuorias, dinámicas de poder e identidad en el cementerio de la misión salesiana Nuestra Señora de la Candelaria (Río Grande, Tierra del Fuego). In V. Aldazábal, L. Amor, M. Díaz, R. Flammini, N. Franco, & B. Matossian (Eds.), Actas de las IV Jornadas Multidisciplinarias (pp. 305–318). Buenos Aires, Argentina: Consejo Nacional de Investigaciones Científicas y Técnicas. Schreyer, A. (2012). Architectural Design with Sketchup, Enhanced Edition: Component-Based Modeling, Plugins, Rendering, and Scripting. Hoboken: John Wiley & Sons Inc. Tal, D. (2009). Google Sketchup for Site Design: A Guide to Modelling Site Plans, Terrain and Architecture. Hoboken: John Wiley & Sons Inc. Tal, D. (2013). Rendering in Sketchup: From Modeling to Presentation for Architecture, Landscape Architecture and Interior Design. Hoboken: Wiley & Sons Inc. Talaverano, R. (2014). Documentación gráfica de edificios históricos: principios, aplicaciones y perspectivas. Arqueología de la Arquitectura, 11, 1–26. http://dx.doi.org/10.3989/arq.arqt.2014.014 Vázquez, F., & Díaz Pais, E. (2014). Arqueología virtual en una estancia colonial argentina. Virtual Archaeology Review, 5(10), 6–10. http://dx.doi.org/10.4995/var.2014.4204

Virtual Archaeology Review, 8(17): 84-94, 2017 94