The Ai First Experience

Total Page:16

File Type:pdf, Size:1020Kb

The Ai First Experience May 23, 2018 THE AI FIRST EXPERIENCE For Internal Use Only - Do Not Distribute CLARIFAI IS A MARKET LEADER IN VISUAL RECOGNITION AI TECHNOLOGY Proven, award-winning technology Team of computer vision experts $40M+ in venture to date Imagenet 2013 Winners Matt Zeiler, CEO & Founder Machine Learning PhD For Internal Use Only - Do Not Distribute COMPUTER VISION RESULTS 90% Accuracy 1980 2013 For Internal Use Only - Do Not Distribute AI 4 NEURAL NETWORK HISTORY 30x Speedup ● more data ● bigger models ● faster iterations 1980s 2009 Geoff Hinton Yann LeCun For Internal Use Only - Do Not Distribute CLARIFAI: A BRIEF HISTORY V2 API Brand Refresh Clarifai launches v2 API with Custom Clarifai launches brand refresh Training and Visual Search On Prem V1 API Series B Clarifai announces On Prem Clarifai launches v1 API Clarifai raises $30M series B funding for image recognition led by Menlo Ventures 2013 2015 2017 2014 2016 2018 Foundation Video Matt Zeiler wins top Clarifai announces video recognition SF Office ImageNet, starts Clarifai capability Clarifai announces opening of SF-based office Series A Clarifai receives $10M series A funding Mobile SDK led by Union Square Ventures Clarifai releases Mobile SDK in Beta Forevery Clarifai launches first consumer app, Forevery photo For Internal Use Only - Do Not Distribute AI 6 POWERFUL ARTIFICIAL INTELLIGENCE IN CLEAN APIS AND SDKS For Internal Use Only - Do Not Distribute AI 7 MODEL GALLERY DEMOS For Internal Use Only - Do Not Distribute AI 8 EASY TO USE API For Internal Use Only - Do Not Distribute AI 9 SIMPLICITY IN USER INTERFACE For Internal Use Only - Do Not Distribute POWERFUL SOLUTIONS FOR EVERY PROBLEM CORE MODEL DOMAIN MODEL CUSTOM MODEL Broad lens Narrow lens Detailed lens DRINK COFFEE STARBUCKS For Internal Use Only - Do Not Distribute CUSTOM TRAINING Teach the AI about the world from your own unique perspective For Internal Use Only - Do Not Distribute Custom Training 12 HOW YOU TRAIN AI Lot of Special Thousands of Data Code Hardware Data Examples Scientists WEEKS For Internal Use Only - Do Not Distribute Custom Training 13 HOW TO GET EVERYONE TO TRAIN AI A few Lines v Hosted Just a Few Individuals Of Code Hardware Data Examples SECONDS For Internal Use Only - Do Not Distribute CUSTOM TRAINING DEMOS Custom Training 15 HOW SIMPLE IS IT TO TRAIN? For Internal Use Only - Do Not Distribute Custom Training 16 SIMPLICITY IN UI For Internal Use Only - Do Not Distribute Demo 17 ORGANIZE WITH AI POWERED SEARCH For Internal Use Only - Do Not Distribute Visual Search Demo 18 SEARCH BY TAG For Internal Use Only - Do Not Distribute Visual Search Demo 19 SEARCH by image red lace up sneakers with weird embellishments For Internal Use Only - Do Not Distribute Visual Search Demo 20 SEARCH by image and tag For Internal Use Only - Do Not Distribute VISUAL SEARCH DEMOS Visual Search Demo 22 HOW SIMPLE IS IT TO SEARCH? For Internal Use Only - Do Not Distribute SIMPLICITY IN USER INTERFACE For Internal Use Only - Do Not Distribute PUSHING AI TO THE EDGE For Internal Use Only - Do Not Distribute LEARN YOUR PREFERENCES For Internal Use Only - Do Not Distribute MONITOR YOUR WORLD For Internal Use Only - Do Not Distribute SEAMLESSLY INTEGRATED Cloud API Mobile SDK On-Premise Integrate machine learning into Run AI on-device with zero Run AI on your servers and your business in no time flat lag, online or offline, to build manage the apps and services with just a few lines of code mobile apps that get more you build on top of it accurate with each use For Internal Use Only - Do Not Distribute PROFIT OPTIMIZATION MONITOR CLARIFAI SOLUTIONS Predict, Train, Search are building blocks to providing real world solutions For Internal Use Only - Do Not Distribute 29 EXAMPLE: FROM DIGITAL TO PHYSICAL - THINKING AI FIRST Bridging the gap between online store experiences and brick & mortar retail. Confidential. For internal use only. 30 THE STORE OF THE FUTURE Practical implementations to... Recommend Analyze & Understand Automate & Maintain Confidential. For internal use only. 31 GUIDE THE BUYER JOURNEY WITH PHOTOS AI enhances shopping of digital catalog from the physical world Confidential. For internal use only. ADD A NEW LAYER OF UNDERSTANDING IN-STORE AI enhances shopping of physical catalog with digital features Confidential. For internal use only. ENHANCE VALUE TO CUSTOMERS WITH AI ROI Method Online In-store Optimize inventory yes yes Organize layout yes yes Profit Optimization New discovery yes yes Recommendation yes yes Analytics yes yes Monitoring Protect Your Brand yes yes Confidential. For internal use only. 34 START BY ANSWERING ONE IMPORTANT QUESTION ARE YOU AN AI COMPANY? NO YES USE A SERVICE BUILD IN-HOUSE For Internal Use Only - Do Not Distribute 35 MACHINE LEARNING EXPERIENCE Max Open-source Max Cost AIaaS Effort level Min Min Min Max Machine Learning Experience For Internal Use Only - Do Not Distribute ACCURACY FOR YOUR PROBLEM Custom Clarifai (algo+data) Custom Clarifai (algo only) Custom Clarifai Algo Improvements Tensorflow “set and forget” General/Domain Clarifai Today Time For Internal Use Only - Do Not Distribute OBJECT DETECTION Detection Aerial Detection For Internal Use Only - Do Not Distribute 38 INDUSTRY Travel USE CASE Auto-tag professional photos from hotels on its platform RESULT • Improved efficiency of human moderators • Curated new content for customers For Internal Use Only - Do Not Distribute 39 INDUSTRY Healthcare SOLUTION Augment doctors with “intelligent” medical diagnoses IMPACT ● 99% accuracy in identifying ear diseases ● Brought modern medicine and specialized ear treatments to remote communities in Africa and south America For Internal Use Only - Do Not Distribute 40 INDUSTRY E-commerce and Retail SOLUTION Automatically recommends relevant West Elm products based on visual similarity to a user’s Pinterest board IMPACT ● Increase basket size ● Increase RPV ● Increase Conversions ● Digital Decor: Buy a room vs. a piece of furniture For Internal Use Only - Do Not Distribute CLARIFAI CUSTOMERS JOIN LEADERS WHO ARE ENABLING AN AI FIRST EXPERIENCE PROFIT OPTIMIZATION MONITOR Recommendation Visual Search Organize & Curate Customer Analytics Moderation Security & Discovery For Internal Use Only - Do Not Distribute THANK YOU For Internal Use Only - Do Not Distribute.
Recommended publications
  • FOR IMMEDIATE RELEASE: Tuesday, April 13, 2021 From: Matthew Zeiler, Clarifai Contact: [email protected]
    FOR IMMEDIATE RELEASE: Tuesday, April 13, 2021 From: Matthew Zeiler, Clarifai Contact: [email protected] https://www.clarifai.com Clarifai to Deliver AI/ML Algorithms on Palantir's Data Management Platform for U.S. Army's Ground Station Modernization Fort Lee, New Jersey - Clarifai, a leadinG AI lifecycle platform provider for manaGinG unstructured imaGe, video and text data, announced today that it has partnered with Palantir, Inc. (NYSE: PLTR) to deliver its Artificial IntelliGence and Machine LearninG alGorithms on Palantir’s data manaGement platform. This effort is part of the first phase of the U.S. Army’s Ground Station modernization in support of its Tactical IntelliGence TargetinG Access Node (TITAN) proGram. Palantir was awarded a Phase one Project Agreement throuGh an Other Transaction Agreement (OTA) with Consortium ManaGement Group, Inc. (CMG) on behalf of Consortium for Command, Control and Communications in Cyberspace (C5). DurinG the first phase of the initiative, Clarifai is supportinG Palantir by providinG object detection models to facilitate the desiGn of a Ground station prototype solution. “By providinG our Computer Vision AI solution to Palantir, Clarifai is helpinG the Army leveraGe spatial, hiGh altitude, aerial and terrestrial data sources for use in intelliGence and military operations,” said Dr. Matt Zeiler, Clarifai’s CEO. “The partnership will offer a ‘turnkey’ solution that inteGrates data from various sources, includinG commercial and classified sources from space to Ground sensors.” “In order to target with Great precision, we need to be able to see with Great precision, and that’s the objective of this proGram,” said Lieutenant General Robert P.
    [Show full text]
  • Scene Understanding for Mobile Robots Exploiting Deep Learning Techniques
    Scene Understanding for Mobile Robots exploiting Deep Learning Techniques José Carlos Rangel Ortiz Instituto Universitario de Investigación Informática Scene Understanding for Mobile Robots exploiting Deep Learning Techniques José Carlos Rangel Ortiz TESIS PRESENTADA PARA ASPIRAR AL GRADO DE DOCTOR POR LA UNIVERSIDAD DE ALICANTE MENCIÓN DE DOCTOR INTERNACIONAL PROGRAMA DE DOCTORADO EN INFORMÁTICA Dirigida por: Dr. Miguel Ángel Cazorla Quevedo Dr. Jesús Martínez Gómez 2017 The thesis presented in this document has been reviewed and approved for the INTERNATIONAL PhD HONOURABLE MENTION I would like to thank the advises and contributions for this thesis of the external reviewers: Professor Eduardo Mario Nebot (University of Sydney) Professor Cristian Iván Pinzón Trejos (Universidad Tecnológica de Panamá) This thesis is licensed under a CC BY-NC-SA International License (Cre- ative Commons AttributionNonCommercial-ShareAlike 4.0 International License). You are free to share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the ma- terial. The licensor cannot revoke these freedoms as long as you follow the license terms. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. NonCommercial — You may not use the material for commercial purposes. ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original. Document by José Carlos Rangel Ortiz. A mis padres. A Tango, Tomás y Julia en su memoria.
    [Show full text]
  • FOR IMMEDIATE RELEASE: Thursday, May 27, 2021 From: Matthew Zeiler, Clarifai Contact: [email protected]
    FOR IMMEDIATE RELEASE: Thursday, May 27, 2021 From: Matthew Zeiler, Clarifai Contact: [email protected] https://www.clarifai.com Clarifai Adds Another Senior Advisor to its Public Sector Advisory Council to Better Serve its Government and Civilian Agency Customers Fort Lee, NJ, May 27, 2021 — Today, Clarifai announced that Jacqueline Tame, the former Acting Deputy Director of the DoD’s Joint Artificial Intelligence Center (JAIC), has Joined the company as a senior advisor and member of its Public Sector Advisory Council. Clarifai is a leading AI provider that specializes in Computer Vision, Natural Language Processing and Audio Recognition. Clarifai helps government and defense agencies transform unstructured image, video, text and audio data into structured data. At the JAIC, Ms. Tame oversaw day-to-day operations, advised DoD AI leadership and led engagements with the White House and Congress to raise awareness of DoD AI programs and shape policy. “Jacqueline Tame has very unique expertise that combines DoD technology policy with implementation of repeatable, outcome-based AI programs” said Dr. Matt Zeiler, Clarifai’s CEO. “She is one of the few people that has successfully worked with Congress to define and fund the US government’s AI policy. We are thrilled to partner with Jacqueline in re-envisioning how our government can work with best-in-class technology companies while harnessing innovation at home.” Ms. Tame has also served as a Senior Advisor to the Under Secretary of Defense for Intelligence and Security; Senior Staffer on the House Permanent Select Committee on Intelligence; Chief of Customer Engagement at the Defense Intelligence Agency; Advisor to the Chief of Naval Operations; and Policy Advisor to the Deputy Director of National Intelligence.
    [Show full text]
  • Challenges and Modern Machine Learning-Based Approaches to Object Tracking in Videos
    Challenges and modern machine learning-based approaches to object tracking in videos Object Tracking Introduction Object tracking, also known as visual tracking, in a video, is one of the challenging tasks in machine vision due to appearance changes caused by variable illumination, occlusion, and variation changes (scale of objects, pose, articulation, etc.). Some of the complex problems in visual tracking is people tracking and traffic density estimation. The conventional algorithms developed by researchers are able to tackle only some of the issues in object tracking, however, they still do not provide a foolproof solution (Bombardelli, Gül, Becker, Schmidt, & Hellge, 2018). Note: It is assumed that readers of this article are well aware of ML techniques used in object detection and object classification. Video tracking aPPlications · Surveillance · Security · Traffic monitoring · Video communication · Robot vision · Accident prediction/detection Can we use object detection at every frame to track an object in a video? This is a very valid question, however, temporal information is a more complex task for non-static items. When things move, there are some occasions when the object disappears for a few frames. Object detection will stop at that point. If the image appears again, a lot of re-processing will be required to identify whether the object that appeared is the same object that was detected previously. The problem becomes more complex when there is more than one object in a video. Object detection is more focused on working with a single image at a time with no idea about motion that may have occurred in the past frame.
    [Show full text]
  • 2018 World AI Industry Development Blue Book 2018 World AI Industry Development Blue Book
    2018 World AI Industry Development Blue Book 2018 World AI Industry Development Blue Book Preface 3 As the driving force of a new round of technology Committed to taking the lead in China’s reform In-depth Report on the and opening-up and innovative developments, Development of the World AI and industrial revolution, artificial intelligence Industry, 2018 (AI) is wielding a profound impact on the world Shanghai has been working closely with economy, social progress and people’s daily life. related national ministries and commissions to Overview and Description Therefore, the 2018 World AI Conference will be jointly organize the 2018 World AI Conference. held in Shanghai from September 17 to 19 this year, Hosting this event becomes a priority in the Industrial Development jointly by the National Development and Reform process of Shanghai constructing the “five Environment Commission (NDRC), the Ministry of Science and centers” - international and economic center, Technology (MST), the Ministry of Industry and finance center, trade center, shipping center Technical Environment Information Technology (MIIT), the Cyberspace and technical innovation center, and building Administration of China (CAC), the Chinese the “four brands” of Shanghai - services, World AI Enterprises Academy of Sciences (CAS), the Chinese Academy manufacturing, shopping and culture. Also, of Engineering (CAE) and the Shanghai Municipal it is the practical measure to foster the reform World Investment and People’s Government, with the approval of the State and opening-up and optimize the business Financing Council, in order to accelerate the development environment in Shanghai. This meeting of a new generation of AI, complying with the provides the opportunity for Shanghai to Industrial Development development rules and grasping the future trends.
    [Show full text]
  • Clarifai Contact: [email protected]
    FOR IMMEDIATE RELEASE: Wednesday, October 28, 2020 From: Matthew Zeiler, Clarifai Contact: [email protected] https://www.clarifai.com Clarifai introduces a new suite of products at its inaugural Perceive 2020 conference NEW YORK, October 28, 2020 — Clarifai, a leading AI lifecycle platform provider for managing unstructured image, video, and text data, hosted its inaugural Perceive 2020 conference on October 21. The conference featured a rock star team of speakers from companies such as Facebook, Google, Samsung, and more. It also included 20 business, technical, panel, and use case sessions. “We hosted this free event to foster a community around AI, where businesses and public sector organizations can learn and collaborate and achieve their mission of accelerating their digital transformation with AI,” said Dr. Matt Zeiler, Clarifai’s CEO and Founder. Clarifai used the event to introduce a suite of new products that position them as the only complete end-to-end AI platform for unstructured data. “Our new Scribe product is a great addition to our end-to- end computer vision and NLP platform. We built it for scalable data labeling so that AI can really learn from accurately annotated datasets,” said Zeiler. In addition to Scribe, Clarifai launched Spacetime — a search tool that can identify when and where objects appear in images and video data. They also launched its Enlight suite for training, managing, and updating deep learning models. Enlight rapidly accelerates the model development process and helps users to iterate on their models over time. They can improve model accuracy and speed while preventing data drift.
    [Show full text]
  • Perspectives on Visual Search Technology What Is Visual Search? Retailers Are Looking for New Ways to Streamline Product Discovery and Visual Search Is Enabling Them
    July 2019 ComCap’s perspectives on Visual Search technology What is Visual Search? Retailers are looking for new ways to streamline product discovery and Visual Search is enabling them ▪ At it’s core, Visual Search can answer key questions that are more easily resolved using visual prompts, unlike traditional text-based searching − From a consumer standpoint, visual search enables product discovery, connecting consumers directly to the products they desire ▪ Visual search is helping blend physical experiences and online convenience as retailers are competing to offer the best service for high-demanding customers through innovative technologies ▪ Consumers are now actively seeking multiple ways to visually capture products across channels, resulting in various applications including: − Image Recognition: enabling consumers to connect pictures with products, either through camera- or online- produced snapshots through AI-powered tools − QR / bar code scanning: allowing quick and direct access between consumers and specific products − Augmented Reality enabled experiences: producing actionable experiences by recognizing objects and layering information on top of a real-time view Neiman Marcus Samsung Google Introduces Amazon Introduces Snap. Introduces Bixby Google Lens Introduces Find. Shop. (2015) Vision (2017) (2017) StyleSnap (2019) Source: eMarketer, News articles 2 Visual Search’s role in retail Visual Search capabilities are increasingly becoming the industry standard for major ecommerce platforms Consumers place more importance on
    [Show full text]
  • Derive Insights from Unstructured Text with NLP
    nATURAL LANGUAGE PROcESSING Derive insights from unstructured text with NLP Get more value out of your text and image data with NLP and Computer Vision. Use these powerful pre-trained models for 3 Topic analysis Classify documents into a language understanding pre-determined set of topics or themes. 1 Text moderation Protect users from toxic, obscene, racist or Smart reply threatening language with pre-trained or 4 custom moderation models. Suggest quick responses for chatbots, emails or other conversational clients. Should this text be shown to our community? Choose the best response: Checking in with you as it’s been too long since we last caught up! Toxic 5 Named entity recognition Sentiment analysis 2 Identify and extract structured text from Assign negative, postive or neutral sentiment documents with the option to add values to documents. bounding boxes. Categorize the sentiment in this text: I’ve been enjoying how easy this product is to use. Founded in 2013 D AT E by Matthew Zeiler NA ME , a foremost expert in machine learning, Clarifai Organization Positive has been a market leader since winning the top five places in image classification at the ImageNet Organization 2013 D AT E competition. optical character recognition Recognize text in images and 98% documents with Object Positive Character Recognition (OCR) Topic Uses of OCR Read text in images and video Extract text from images to gain more meaning and metadata from your content. Text name Metadata information Process documents at scale Why Clarifai Process documents automatically and at scale with the power of Computer Vision. Clarifai is the leading enterprise platform for Computer Vision, Natural Language Processing and deep learning to model unstructured image, video and textual data.
    [Show full text]
  • Review of Deep Convolution Neural Network in Image Classification
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by UMP Institutional Repository 2017 International Conference on Radar, Antenna, Microwave, Electronics, and Telecommunications Review of Deep Convolution Neural Network in Image Classification Ahmed Ali Mohammed Al-Saffar, Hai Tao, Mohammed Ahmed Talab Faculty of Computer Systems and Software Engineering Universiti Malaysia Pahang Pahang, Malaysia [email protected] Abstract—With the development of large data age, recognition system, feature extraction methods such as Scale- Convolutional neural networks (CNNs) with more hidden layers invariant feature transform (SIFT [1]) and histogram of have more complex network structure and more powerful feature oriented gradients (HOG [2]) were used, and then the extracted learning and feature expression abilities than traditional machine Feature input classifier for classification and recognition. These learning methods. The convolution neural network model trained features are essentially a feature of manual design. For by the deep learning algorithm has made remarkable different identification problems, the extracted features have a achievements in many large-scale identification tasks in the field direct impact on the performance of the system, so the of computer vision since its introduction. This paper first researchers need to study the problem areas to be studied in introduces the rise and development of deep learning and order to design Adaptability to better features, thereby convolution neural network,
    [Show full text]
  • The Malicious Use of Artificial Intelligence: February 2018 Forecasting, Prevention, and Mitigation
    Future of University Centre for University of Center for a Electronic OpenAI Humanity of Oxford the Study of Cambridge New American Frontier Institute Existential Security Foundation Risk The Malicious Use February 2018 of Artificial Intelligence: Forecasting, Prevention, and Mitigation The Malicious Use of Artificial Intelligence: February 2018 Forecasting, Prevention, and Mitigation Miles Brundage Shahar Avin Jack Clark Helen Toner Peter Eckersley Ben Garfinkel Allan Dafoe Paul Scharre Thomas Zeitzoff Bobby Filar Hyrum Anderson Heather Roff Gregory C. Allen Jacob Steinhardt Carrick Flynn Seán Ó hÉigeartaigh Simon Beard Haydn Belfield Sebastian Farquhar Clare Lyle Rebecca Crootof Owain Evans Michael Page Joanna Bryson Roman Yampolskiy Dario Amodei 1 Corresponding author 9 American University 18 Centre for the Study of miles.brundage@philosophy. Existential Risk, University ox.ac.uk of Cambridge Future of Humanity Institute, 10 Endgame University of Oxford; Arizona State University 19 Future of Humanity 11 Endgame Institute, University of 2 Corresponding author, Oxford [email protected] University of Oxford/ Centre for the Study of 12 Arizona State University/New Existential Risk, University 20 Future of Humanity America Foundation of Cambridge Institute, University of Oxford 13 Center for a New American 3 OpenAI Security 21 Information Society Project, Yale University Open Philanthropy Project 4 14 Stanford University 22 Future of Humanity Electronic Frontier Institute, University of 5 Future of Humanity Foundation 15 Oxford Institute,
    [Show full text]
  • A State-Of-The-Art Survey on Deep Learning Theory and Architectures
    Review A State-of-the-Art Survey on Deep Learning Theory and Architectures Md Zahangir Alom 1,*, Tarek M. Taha 1, Chris Yakopcic 1, Stefan Westberg 1, Paheding Sidike 2, Mst Shamima Nasrin 1, Mahmudul Hasan 3, Brian C. Van Essen 4, Abdul A. S. Awwal 4 and Vijayan K. Asari 1 1 Department of Electrical and Computer Engineering, University of Dayton, OH 45469, USA; [email protected] (T.M.T.); [email protected] (C.Y.); [email protected] (S.W.); [email protected] (M.S.N.); [email protected] (V.K.A.) 2 Department of Earth and Atmospheric Sciences, Saint Louis University, MO 63108, USA; [email protected] 3 Comcast Labs, Washington, DC 20005, USA; [email protected] 4 Lawrence Livermore National Laboratory (LLNL), Livermore, CA 94550, USA; [email protected] (B.C.V.E.); [email protected] (A.A.S.A.) * Correspondence: [email protected] Received: 17 January 2019; Accepted: 31 January 2019; Published: 5 March 2019 Abstract: In recent years, deep learning has garnered tremendous success in a variety of application domains. This new field of machine learning has been growing rapidly and has been applied to most traditional application domains, as well as some new areas that present more opportunities. Different methods have been proposed based on different categories of learning, including supervised, semi-supervised, and un-supervised learning. Experimental results show state-of-the- art performance using deep learning when compared to traditional machine learning approaches in the fields of image processing, computer vision, speech recognition, machine translation, art, medical imaging, medical information processing, robotics and control, bioinformatics, natural language processing, cybersecurity, and many others.
    [Show full text]
  • A Multi-Modal Approach to Infer Image Affect
    A Multi-Modal Approach to Infer Image Aect Ashok Sundaresan, Sugumar Murugesan Sean Davis, Karthik Kappaganthu, ZhongYi Jin, Divya Jain Anurag Maunder Johnson Controls - Innovation Garage Santa Clara, CA ABSTRACT Our approach combines the top-down and boom-up features. e group aect or emotion in an image of people can be inferred Due to the presence of multiple subjects in an image, a key problem by extracting features about both the people in the picture and the that needs to be resolved is to combine individual human emotions overall makeup of the scene. e state-of-the-art on this problem to a group level emotion. e next step is to combine multiple investigates a combination of facial features, scene extraction and group level modalities. We address the rst problem by using a Bag- even audio tonality. is paper combines three additional modali- of-Visual-Words (BOV) based approach. Our BOV based approach ties, namely, human pose, text-based tagging and CNN extracted comprises developing a code book using two well known clustering features / predictions. To the best of our knowledge, this is the algorithms: k-means clustering and a Gaussian mixture model rst time all of the modalities were extracted using deep neural (GMM) based clustering. e code books from these clustering networks. We evaluate the performance of our approach against algorithms are used to fuse individual subjects’ features to a group baselines and identify insights throughout this paper. image level features We evaluate the performance of individual features using 4 dif- KEYWORDS ferent classication algorithms, namely, random forests, extra trees, gradient boosted trees and SVM.
    [Show full text]