DOCSLIB.ORG

Shadow Map and Ambient Occlusion

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text

Load more

Shadow Mapping A point is lit if it is “visible” EECS 487: Interactive from the light source • similar to visible surface Computer Graphics determination Shadow computation by Lecture 32: Interactive Visual Effects simulating eye at light • Shadow Map position • Ambient Occlusion Durand Shadow Mapping Shadow Mapping Requires 2 passes through the pipeline First pass: compute shadow map Second pass: shadow determination (depth of closest pixels to the light) using the shadow map • render scene from light For each pixel • populate z-buffer that we’ll use as our shadow map • do normal z-buffer computation • store the distance from light to nearest object to check visibility from eye • white is far, black is near shadow map • if visible, look up distance to light scene from light’s point of view • perspective project visible pixels from eye scene from eye’s point of view depth value space back to light space • lookup depth stored in shadow map • if distance to light is (epsilon) greater than stored depth, pixel is in shadow Foley et al., Akenine-Möller02,Durand Foley et al., Akenine-Möller02,Durand Shadow Mapping with OpenGL Shadow Mapping with OpenGL Create the shadow map: Determine eye-space, light-space correspondences: • render the scene with eye at light position • unproject every pixel in the z-buffer to obtain object • save the resulting z-buffer and projection matrix, coordinates, using gluUnproject(…) and the these are the shadow map and shadow projection matrix viewpoint projection matrix • glReadPixels(…) to read back z-buffer • glGetDoublev(GL_MODELVIEW_MATRIX, …), • project the object coordinates into the shadow map using glGetDoublev(GL_PROJECTION_MATRIX, …), gluProject(…) and the shadow projection matrix glGetIntergerv(GL_VIEWPORT, …) Render scene: Shadow rendering: • compare the resulting z values with the corresponding • render the scene from the actual viewpoint content of the shadow map and update the color buffer to • save both color and z-buffers and viewpoint projection matrix draw the projected shadows • write back the modified color buffer using glDrawPixels(…) Yu Yu Limited Field of View Limitations of Shadow Maps What if a point/ is outside field of 1. Limited field of view, no omni-directional light view of the shadow map? 2. Limited depth resolution Use six shadow maps, to form a cube 3. Shadow map aliasing enclosing the light • requires a separate rendering 4. Hard shadows only (with jaggies) pass for each shadow map! (or soft shadows only if PCF is used) Durand Durand Limited Depth Resolution Shadow Map Aliasing Due to z-fighting, distance to light and stored depth may not compare correctly Results from sampling rates • add an ε to depth in shadow map mismatch (re-sampling to prevent unintended (self-)shadowing problem): sampling rate (pixel • in computing shadow map move geometry away from light by a small amount size) of light is mismatched to • choosing correct ε value is tricky that of the eye ε precision Precision error also error possible with projection shadows Merrell Kilgard Shadow Map Aliasing Shadow Map Aliasing Shadow “fragment” is also Least aliasing when light frustum is reasonably stretched when eye is close to well aligned with the eye’s view frustum: the the surface but light is far away ratio of sample sizes is close to 1 • surfaces that are nearly edge-on to the eye face the light directly reference image • best case if eye and light frusta are nearly identical • results in under-sampling of near (“miner’s lamp” case) field in shadow map • but only limited scene setups satisfy this • worst case is when light is shining at the viewer (“deer-in-the-headlights” case) • also known as the “dueling frusta” problem projected fragment Kilgard Durand,Merrell,RTR3 standard shadow map Shadow Map Anti-Aliasing Shadow Map Anti-Aliasing Possible solutions: Asymmetric frustrum for shadow map rendering 1. increase shadow map resolution 2. split shadow map into several slices different projections, same resolution 3. use asymmetric frustrum for shadow map rendering 4. average nearby pixels Durand,RTR3 RTR3,Martin&Tan Percentage-Closer Filtering Soft-Shadows with Shadow Map depth values Look up several nearby shadow map fragments, not just one Compute average shadow value for the neighborhood If so, (5/9) • use immediate neighbors for anti-aliasing occludes • surface use neighbors further afield for soft-shadows Sampling strategy: grid, jittered, or adaptive for Results in aliased shadow improved performance Anti-aliasing by averaging several nearby shadow map fragments What to average? Depth values? • No, 1.2 < 49.8, but so is 22.9: still a binary result, no anti-aliased blurring Instead, perform depth test for a neighborhood of pixels • then compute percentage of lit pixels • (this makes ε computation even trickier!) Durand,Schulze Percentage-Closer Filtering Ambient Occlusion Supported in hardware for small (2×2) filters All the real-time shadow algorithms assume directional lighting only • shadow map coordinates generated using projection matrix Ambient occlusion is shadows due to • shadow map stored as texture: modern hardware permits tests on texture values global indirect/ambient lighting • can use larger filters with additional rendering passes Can be pre-computed using global illumination and baked into a texture limited to static scenes Yang et al., Durand,Schulze TP3 Ambient Occlusion SSAO Two approaches: • Screen-space Ambient Occlusion (SSAO) At each point find fraction of • Screen-space Directional Occlusion (SSDO) hemisphere that is occluded • visible fraction: 1-occlusion Basic idea: • modulate diffuse shading by • approximate indirect lighting using a uniform, (1−occlusion) distant environment irradiance • cd = md sd max((n•l), 0) (1−occlusion) • simulate the darkening effect where ambient light is blocked by other geometry in the scene Alternately, to avoid ray tracing, • compute the portion of the hemisphere around a point that is blocked sample a circle around point • purely geometric, independent of lighting • compare depth of samples conditions or viewing direction against z-buffer content • sample is not occluded if depth < z-buffer content −Z TP3 Bala, Yang, Yip, Xu n SSDO b Shadow and Environment Maps SSAO plus EM and indirect lighting: p • instead of modulating diffuse shading by (1−occlusion), Basic method to add realism to interactive rendering: add up light contribution from unoccluded directions only instead of ray tracing, use image-based methods • compute average un-occluded direction (“bent normal”, b) • Shadow maps: image-based hard shadows [Williams78] • lookup light/environment map using the “bent normal” • many recent extensions result in light-colored shadow instead of just grey • widely used even in software rendering (RenderMan) • (can add a single bounce off facing occluding surfaces) • Environment maps: image-based complex lighting Alternately, to avoid ray tracing, [Blinn&Newell76] • sample uniformly across the hemisphere • huge amount of recent work above point, this gives N light vectors • sample each light vector at random offsets from the point Together, give many “realistic” effects • compare depth of samples against z-buffer content • sample is not occluded if depth < z-buffer content • but cannot be easily combined! Ritschel,Grosch,Seidel09 Ramamoorthi .
Recommended publications
  • Raytracing Prefiltered Occlusion for Aggregate Geometry

    Raytracing Prefiltered Occlusion for Aggregate Geometry

    Raytracing Prefiltered Occlusion for Aggregate Geometry Dylan Lacewell1,2 Brent Burley1 Solomon Boulos3 Peter Shirley4,2 1 Walt Disney Animation Studios 2 University of Utah 3 Stanford University 4 NVIDIA Corporation Figure 1: Computing shadows using a prefiltered BVH is more efficient than using an ordinary BVH. (a) Using an ordinary BVH with 4 shadow rays per shading point requires 112 seconds for shadow rays, and produces significant visible noise. (b) Using a prefiltered BVH with 9 shadow rays requires 74 seconds, and visible noise is decreased. (c) Reducing noise to a similar level with an ordinary BVH requires 25 shadow rays and 704 seconds (about 9.5× slower). All images use 5 × 5 samples per pixel. The scene consists of about 2M triangles, each of which is semi-opaque (α = 0.85) to shadow rays. ABSTRACT aggregate geometry, it suffices to use a prefiltering algorithm that is We prefilter occlusion of aggregate geometry, e.g., foliage or hair, linear in the number of nodes in the BVH. Once built, a prefiltered storing local occlusion as a directional opacity in each node of a BVH does not need to be updated unless geometry changes. bounding volume hierarchy (BVH). During intersection, we termi- During rendering, we terminate shadow rays at some level of the nate rays early at BVH nodes based on ray differential, and compos- BVH, dependent on the differential [10] of each ray, and return the ite the stored opacities. This makes intersection cost independent of stored opacity at the node. The combined opacity of the ray is com- geometric complexity for rays with large differentials, and simulta- puted by compositing the opacities of one or more nodes that the neously reduces the variance of occlusion estimates.
  • Efficient Image-Based Methods for Rendering Soft Shadows

    Efficient Image-Based Methods for Rendering Soft Shadows

    Efficient Image-Based Methods for Rendering Soft Shadows Maneesh Agrawala Ravi Ramamoorthi Alan Heirich Laurent Moll Pixar Animation Studios Stanford University∗ Compaq Computer Corporation Figure 1: A plant rendered using our interactive layered attenuation-map approach (left), rayshade (middle), and our efficient high-quality coherence-based raytracing approach (right). Note the soft shadows on the leaves. To emphasize the soft shadows, this image is rendered without cosine falloff of light intensity. Model courtesy of O. Deussen, P. Hanrahan, B. Lintermann, R. Mech, M. Pharr, and P. Prusinkiewicz. Abstract The cost of many soft shadow algorithms grows with the geomet- ric complexity of the scene. Algorithms such as ray tracing [5], We present two efficient image-based approaches for computation and shadow volumes [6], perform visibility calculations in object- and display of high-quality soft shadows from area light sources. space, against a complete representation of scene geometry. More- Our methods are related to shadow maps and provide the associ- over, some interactive techniques [12, 27] precompute and display ated benefits. The computation time and memory requirements for soft shadow textures for each object in the scene. Such approaches adding soft shadows to an image depend on image size and the do not scale very well as scene complexity increases. number of lights, not geometric scene complexity. We also show Williams [30] has shown that for computing hard shadows from that because area light sources are localized in space, soft shadow point light sources, a complete scene representation is not nec- computations are particularly well suited to image-based render- essary.
  • Fast Precomputed Ambient Occlusion for Proximity Shadows

    Fast Precomputed Ambient Occlusion for Proximity Shadows

    INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET EN AUTOMATIQUE Fast Precomputed Ambient Occlusion for Proximity Shadows Mattias Malmer — Fredrik Malmer — Ulf Assarsson — Nicolas Holzschuch N° 5779 Décembre 2005 Thème COG apport de recherche ISRN INRIA/RR--5779--FR+ENG ISSN 0249-6399 Fast Precomputed Ambient Occlusion for Proximity Shadows Mattias Malmer∗ , Fredrik Malmer∗ , Ulf Assarsson† ‡ , Nicolas Holzschuch‡ Thème COG — Systèmes cognitifs Projets ARTIS Rapport de recherche n° 5779 — Décembre 2005 — 19 pages Abstract: Ambient occlusion is used widely for improving the realism of real-time lighting simulations. We present a new, simple method for storing ambient occlusion values, that is very easy to implement and uses very little CPU and GPU resources. This method can be used to store and retrieve the percentage of occlusion, either alone or in combination with the average occluded direction. The former is cheaper in memory costs, while being slightly less accurate. The latter is slightly more expensive in memory, but gives more accurate results, especially when combining several occluders. The speed of our algorithm is independent of the complexity of either the occluder or the receiving scene. This makes the algorithm highly suitable for games and other real-time applications. Key-words: ∗ Syndicate Ent., Grevgatan 53, SE-114 58 Stockholm, Sweden. † Chalmers University of Technology, SE-412 96 Göteborg, Sweden. ‡ ARTIS/GRAVIR – IMAG INRIA Rhône-Alpes, France. Unité de recherche INRIA Rhône-Alpes 655, avenue de l’Europe, 38334 Montbonnot Saint Ismier (France) Téléphone : +33 4 76 61 52 00 — Télécopie +33 4 76 61 52 52 Fast Precomputed Ambient Occlusion for Proximity Shadows Résumé : L’Ambient Occlusion est fréquemment utilisée pour améliorer le réalisme de simulations de l’éclairage en temps-réel.
  • Shadows in Computer Graphics

    Shadows in Computer Graphics

    Shadows in Computer Graphics by Björn Kühl im/ve University of Hamburg, Germany Importance of Shadows Shadows provide cues − to the position of objects casting and receiving shadows − to the position of the light sources Importance of Shadows Shadows provide information − about the shape of casting objects − about the shape of shadow receiving objects Shadow Tutorial Popular Shadow Techniques Implementation Details − Stencil Shadow Volumes (object based) − Shadow Mapping (image based) Comparison of both techniques Popular Shadow Methods Shadows can be found in most new games Two methods for shadow generation are predominately used: − Shadow Mapping Need for Speed Prince of Persia − Stencil Shadow Volumes Doom3 F.E.A.R Prey Stencil Shadow Volumes Figure: Screenshot from id's Doom3 Stencil Shadow Volumes introduction Frank Crow introduced his approach using shadow volumes in 1977. Tim Heidmann of Silicon Graphics was the first one who implemented Crow´s idea using the stencil buffer. Figure: The shadow volume encloses the region which could not be lit. Stencil Shadow Volumes Calculating Shadow Volumes With The CPU Calculate the silhouette: An edge between two planes is a member of the silhouette, if one plane is facing the light and the other is turned away. Figure: silhouette edge (left), a light source and an occluder (top right), and the silhouette of the occluder ( down right) Stencil Shadow Volumes Calculating Shadow Volumes With The CPU The shadow volume should be closed By extruding the silhouette, the side surfaces of the shadow volume are generated The light cap is formed by the planes facing the light The dark cap is formed by the planes not facing the light − which are transferred into infinity.
  • GEARS: a General and Efficient Algorithm for Rendering Shadows

    GEARS: a General and Efficient Algorithm for Rendering Shadows

    Lili Wang et al. / GEARS: A General and Efficient Algorithm for Rendering Shadows 1 GEARS: A General and Efficient Algorithm for Rendering Shadows Lili Wang†1, Shiheng Zhou1, Wei Ke2, and Voicu Popescu3 1China State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing China 2 Macao Polytechnic Institute, Macao China 3Computer Science, Purdue University, West Lafayette, Indiana, USA Figure 1: Soft shadows rendered with our method (left of each pair) and with ray tracing (right of each pair), for comparison. The average frame rate for our method vs. ray tracing is 26.5fps vs. 1.07fps for the Bird Nest scene, 75.6fps vs. 5.33fps for the Spider scene, and 21.4fps vs. 1.98fps for the Garden scene. The shadow rendering times are 49.8ms vs. 1176.4ms, 14.3ms vs. 209.2ms, and 45.2ms vs. 561.8ms for the three scenes, respectively. The output image resolution is 512x512. submitted to COMPUTER GRAPHICS Forum (3/2014). Volume xx (200y), Number z, pp. 1–13 Abstract We present a soft shadow rendering algorithm that is general, efficient, and accurate. The algorithm supports fully dynamic scenes, with moving and deforming blockers and receivers, and with changing area light source parameters. For each output image pixel, the algorithm computes a tight but conservative approximation of the set of triangles that block the light source as seen from the pixel sample. The set of potentially blocking triangles allows estimating visibility between light points and pixel samples accurately and efficiently. As the light source size decreases to a point, our algorithm converges to rendering pixel accurate hard shadows.
  • 1 Camera Space Shadow Maps for Large Virtual Environments

    1 Camera Space Shadow Maps for Large Virtual Environments

    Camera Space Shadow Maps for Large Virtual Environments Ivica Kolic 1 ⋅⋅⋅ Zeljka Mihajlovic 2 Abstract This paper presents a new single-pass shadow mapping technique that achieves better 1 Introduction quality than the approaches based on perspective warping, such as perspective, light-space and A shadow is one of the most important elements for trapezoidal shadow maps. The proposed technique achieving realism in virtual environments. Over the is appropriate for real-time rendering of large years, many real-time shadow-generation virtual environments that include dynamic objects. approaches have been developed. The most well- By performing operations in camera space, this known approaches are fake and planar shadows, solution successfully handles the general and the shadow volumes (Crow 1977 ), and shadow dueling frustum cases, and produces high-quality mapping (Williams 1978 ). shadows even for extremely large scenes. This This paper primarily focuses on the shadow paper also presents a fast non-linear projection mapping technique that is recognized as the most technique for shadow map stretching that enables important shadow generation tool today because of complete utilization of the shadow map by its simplicity, generality, and predictability. The eliminating wastage. The application of stretching basic shadow mapping principle involves two results in a significant reduction in unwanted steps. In the first step, the shadow map is generated perspective aliasing, commonly found in all by storing the depth of the scene from the point of shadow mapping techniques. Technique is view of the light. In the second step, the scene is compared with other shadow mapping techniques, rendered regularly; however, for every drawn pixel, and the benefits of the proposed method are the shadow map is consulted to determine whether presented.
  • Improving Shadows and Reflections Via the Stencil Buffer

    Improving Shadows and Reflections Via the Stencil Buffer

    Improving Shadows and Reflections via the Stencil Buffer Mark J. Kilgard * NVIDIA Corporation “Slap textures on polygons, throw them at the screen, and let the depth buffer sort it out.” For too many game developers, the above statement sums up their approach to 3D hardware- acceleration. A game programmer might turn on some fog and maybe throw in some transparency, but that’s often the limit. The question is whether that is enough anymore for a game to stand out visually from the crowd. Now everybody’s using 3D hardware-acceleration, and everybody’s slinging textured polygons around, and, frankly, as a result, most 3D games look more or less the same. Sure there are differences in game play, network interaction, artwork, sound track, etc., etc., but we all know that the “big differentiator” for computer gaming, today and tomorrow, is the look. The problem is that with everyone resorting to hardware rendering, most games look fairly consistent, too consistent. You know exactly the textured polygon look that I’m talking about. Instead of settling for 3D hardware doing nothing more than slinging textured polygons, I argue that cutting-edge game developers must embrace new hardware functionality to achieve visual effects beyond what everybody gets today from your basic textured and depth-buffered polygons. Two effects that can definitely set games apart from the crowd are better quality shadows and reflections. While some games today incorporate shadows and reflections to a limited extent, the common techniques are fairly crude. Reflections in today’s games are mostly limited to “infinite extent” ground planes or ground planes bounded by walls.
  • Combining Screen-Space Ambient Occlusion and Cartoon Rendering on Graphics Hardware

    Combining Screen-Space Ambient Occlusion and Cartoon Rendering on Graphics Hardware

    Combining Screen-Space Ambient Occlusion and Cartoon Rendering on Graphics Hardware Brett Lajzer Dan Nottingham Figure 1: Four visualizations of the same scene: a) no SSAO or outlining, b) SSAO, no outlines, c) no SSAO, outlines, d) SSAO and outlines 1. Motivation Screen-Space Ambient Occlusion Screen-space ambient occlusion (SSAO) is a further Methods for non-photorealistic rendering of 3D scenes have approximation of this technique, which was developed by become more popular in recent years for computer CryTek for their game Crysis and its engine. This version animation and games. We were interested in combining computes ambient occlusion for each pixel visible on the two particular NPR techniques: ambient occlusion and screen, by generating random points in the hemisphere cartoon shading. Ambient occlusion is an approach to around that pixel, and determining occlusion for each point global lighting that assumes that a point on the surface of by comparing its depth to a depth map of the scene. The an object receives less ambient light if there are many other sample is considered occluded if it is further from the objects occupying the space nearby in the hemisphere camera than the depth of the nearest visible object at that around that point. Screen-space ambient occlusion point, unless the difference in depth is greater than the approximates this on the GPU using the depth buffer to test sample radius. The advantage of this method is that it can occlusion of sample points. We combine this with cartoon be implemented on graphics hardware and run in real time, shading, which draws dark outlines on objects based on making it more suited to dynamic, interactive scenes due to depth and normal discontinuities, and thresholds lighting its dependence only upon screen resolution rather than intensity to several discreet values.
  • Fast Image-Based Ambient Occlusion IBAO

    Fast Image-Based Ambient Occlusion IBAO

    The International Journal of Virtual Reality, 2011, 10(4):61-65 61 Fast Image-Based Ambient Occlusion IBAO Robert Sajko and Zeljka Mihajlovic University of Zagreb, Faculty of Electrical Engineering and Computing,Department of Electronics, Microelectronics, Computer and Intelligent Systems Ambient lighting is an approximation of the light reflected from Abstract— The quality of computer rendering and perception other objects in the scene. Its presence reveals the spatial of realism greatly depend on the shading method used to relationship between objects, their shape, depth and surface implement the interaction of light with the surfaces of objects in a complexity details. Ambient lighting can be locally occluded scene. Ambient occlusion (AO) enhances the realistic impression by nearby object or a fold in the surface. Ambient occlusion of rendered objects and scenes. Properties that make Screen produces only subtle visual cues, however, they are very Space Ambient Occlusion (SSAO) interesting for real-time important in natural perception and thus also in a convincing, graphics are scene complexity independence, and support for fully dynamic scenes. However, there are also important issues with realistic lighting model. current approaches: poor texture cache use, introduction of noise, Modern consumer GPUs offer impressive computational and performance swings. power which has allowed the use of various techniques and In this paper, a straightforward solution is presented. Instead algorithms in real-time computer graphics that were previously of a traditional, geometry-based sampling method, a novel, possible in offline rendering only. One such technique is image-based sampling method is developed, coupled with a ambient occlusion, which approximates soft shadows due to revised heuristic function for computing occlusion.
  • Neural Network Ambient Occlusion

    Neural Network Ambient Occlusion

    Neural Network Ambient Occlusion Daniel Holden∗ Jun Saitoy Taku Komuraz University of Edinburgh Method Studios University of Edinburgh Figure 1: Comparison showing Neural Network Ambient Occlusion enabled and disabled inside a game engine. Abstract which is more accurate than existing techniques, has better perfor- mance, no user parameters other than the occlusion radius, and can We present Neural Network Ambient Occlusion (NNAO), a fast, ac- be computed in a single pass allowing it to be used as a drop-in curate screen space ambient occlusion algorithm that uses a neural replacement for existing techniques. network to learn an optimal approximation of the ambient occlu- sion effect. Our network is carefully designed such that it can be 2 Related Work computed in a single pass allowing it to be used as a drop-in re- placement for existing screen space ambient occlusion techniques. Screen Space Ambient Occlusion Screen Space Ambient Oc- clusion (SSAO) was first introduced by Mittring [2007] for use in Keywords: neural networks, machine learning, screen space am- Cryengine2. The approach samples around the depth buffer in a bient occlusion, SSAO, HBAO view space sphere and counts the number of points which are in- side the depth surface to estimate the occlusion. This method has seen wide adoption but often produces artifacts such as dark halos 1 Introduction around object silhouettes or white highlights on object edges. Fil- ion and McNaughton [2008] presented SSAO+, an extension which Ambient Occlusion is a key component in the lighting of a scene samples in a hemisphere oriented in the direction of the surface nor- but expensive to calculate.
  • Shadow Mappingmapping

    Shadow Mappingmapping

    ShadowShadow MappingMapping Cass Everitt NVIDIA Corporation [email protected] Motivation for Better Shadows • Shadows increase scene realism • Real world has shadows • Other art forms recognize the value of shadows • Shadows provide important depth cues 2 Common Real-time Shadow Techniques Projected Projected Shadow planar planar volumes shadows Hybrid approaches Light maps 3 Problems with Common Shadow Techniques • Mostly tricks with lots of limitations • Projected planar shadows • well works only on flat surfaces • Stenciled shadow volumes • determining accurate, water-tight shadow volume is hard work • Light maps • totally unsuited for dynamic shadows • In general, hard to get everything shadowing everything 4 Stenciled Shadow Volumes • Powerful technique, excellent results possible 5 Introducing Another Technique: Shadow Mapping • Image-space shadow determination • Lance Williams published the basic idea in 1978 • By coincidence, same year Jim Blinn invented bump mapping (a great vintage year for graphics) • Completely image-space algorithm • means no knowledge of scene’s geometry is required • must deal with aliasing artifacts • Well known software rendering technique • Pixar’s RenderMan uses the algorithm • Basic shadowing technique for Toy Story, etc. 6 Shadow Mapping References • Important SIGGRAPH papers • Lance Williams, “Casting Curved Shadows on Curved Surfaces,” SIGGRAPH 78 • William Reeves, David Salesin, and Robert Cook (Pixar), “Rendering antialiased shadows with depth maps,” SIGGRAPH 87 • Mark Segal, et. al. (SGI), “Fast
  • Reflective Shadow Maps

    Reflective Shadow Maps

    Reflective Shadow Maps Carsten Dachsbacher∗ Marc Stamminger† University of Erlangen-Nuremberg University of Erlangen-Nuremberg Figure 1: This figure shows the components of the reflective shadow map (depth, world space coordinates, normal, flux) and the resulting image rendered with indirect illumination from the RSM. Note that the angular decrease of flux is shown exaggerated for visualization. Abstract mination, however, generates subtle, but also important effects that are mandatory to achieve realism. In this paper we present ”reflective shadow maps”, an algorithm for Unfortunately, due to their global nature, full global illumina- interactive rendering of plausible indirect illumination. A reflective tion and interactivity are usually incompatible. Ray Tracing and shadow map is an extension to a standard shadow map, where every Radiosity—just to mention the two main classes of global illumi- pixel is considered as an indirect light source. The illumination due nation algorithms—require minutes or hours to generate a single to these indirect lights is evaluated on-the-fly using adaptive sam- image with full global illumination. Recently, there has been re- pling in a fragment shader. By using screen-space interpolation of markable effort to make ray tracing interactive (e.g. [Wald et al. the indirect lighting, we achieve interactive rates, even for complex 2003]). Compute clusters are necessary to achieve interactivity at scenes. Since we mainly work in screen space, the additional effort good image resolution and dynamic scenes are difficult to handle, is largely independent of scene complexity. The resulting indirect because they require to update the ray casting acceleration struc- light is approximate, but leads to plausible results and is suited for tures for every frame.