Research at Google: With a Case of YouTube-8M. Joonseok Lee, Google Research
|
|
- Ezra Heath
- 5 years ago
- Views:
Transcription
1 Research at Google: With a Case of YouTube-8M Joonseok Lee, Google Research
2 Google Research We tackle the most challenging problems in Computer Science and related fields. Being bold and taking risks allows our embedded teams to make discoveries that affect billions of users every day. Ever since Google was born in Stanford's Computer Science department, the company has valued and maintained strong relations with universities and research institutes. 2
3 Today s Talk: YouTube-8M Video Challenging Understanding Problem Problem Large-Scale Billions of Challenges Users Kaggle Academic Competition Relationship + and CVPR 2017 Contribution Workshop 3
4 Video Understanding Problem 4
5 What is Video Understanding? From raw signals... to meaning... useful for Annotation Classification Recommendation Search Summarization... 5
6 The Multiple Shades of Video Understanding kids playing park cars honking sidewalk protagonist policeman shouting stop! Police Chase intro indoor dialog poorly lit outdoor chase hand-held camera Describing the content: what is visible/audible? Inferring the central topics: what is the story about? Describing the structure & style: how is the story told? credits Inferring creator / viewer intent: why capture this video? why watch this video? 6
7 Applications: YouTube Video Discovery Content Metadata Fuser Viewer signals YouTube Auto-generated Channel Topics Topic annotations describe videos and channels 7
8 Applications: Personal Media Collections Lots of videos No metadata 8
9 Applications: Cloud Video Intelligence API Insite from Videos Cloud Video Intelligence API allows developers to extract actionable insights from video files without requiring any machine learning or computer vision knowledge. 9
10 Large-Scale Challenges 10
11 Challenges in Creating Video Dataset File sizes are larger than images. Video labels are more expensive to obtain. More expensive to download, store, and train from. Requiring annotators to watch the video and listen to audio stream. Therefore, video datasets tend to be small. 11
12 YouTube-8M: What is it? Dataset & open-source TensorFlow code research.google.com/youtube8m/ github.com/google/youtube-8m/ 1 Petabyte of data served so far! Kaggle Competition (2017/2/ /6/2) kaggle.com/c/youtube8m $100,000 prize pool (sponsored by Cloud ML) $30,000 in Cloud credits for participants CVPR 17 Workshop (2017/7/26) research.google.com/youtube8m/workshop.html 4 invited talks, 10 oral + 8 poster presentations 12
13 research.google.com/youtube8m/ YouTube-8M Dataset: Vocabulary 4,716 Knowledge Graph entities, each entity has 200+ corresponding videos 13
14 YouTube-8M: Diversity 14
15 YouTube-8M: TensorFlow Framework Design YT-8M (original videos) HMDB Video/Audio Feature Extraction UCF101 Computation per example ImageNet YT-8M (pre-computed features) MNIST Data Size github.com/google/youtube-8m/ Large data size and lower compute intensity 15
16 Kaggle Competition & CVPR Workshop 16
17 The YouTube-8M Classification Challenge Input: Target: A sequence of frame-level visual and audio features, extracted at 1 frame-per-second Each video has Visual Inception-V3 bottleneck features extracted from pixels (PCA-ed to 1024-d) Audio VGG-style bottleneck features extracted from audio spectrograms (128-d) Video topics from a 4,716 Knowledge Graph entity vocabulary The target topics cover the main themes in the video (vs. object detection, scene parsing) Each video has 3.4 ground truth labels on average Goal: Predict target video topics from the sequence of frame-level features 17
18 The YouTube-8M Classification Challenge Korean Food Cooking Meat Football Machine Learning Model Feature Feature Feature Feature 18
19 Participation Statistics: Overall Submissions Received: 7,833 (73.2/day in average) Unique Page Views: 145,863 Downloaders: 3,024 Competing users: 926 Leaderboard top score Competing teams:
20 Number of Submissions 20
21 Where are the participants from? Participants from 56 countries Number of submissions USA: 2,810 China: 1,675 Korea: 420 UK: 370 Number of participants USA: 293 China: 94 India: 48 Russia: 32 Korea, UK: 31 21
22 Final standing: Top 500 The top-performing team (84.97%, rank 1) LSTM starter code baseline (80.93%, rank 78) Audio-visual MoE video-level baseline (~78%) Audio-visual log-reg baseline (74.71%) Visual log-reg baseline (69.42%) 22
23 Final standing: Top 20 INRIA Tsinghua University Baidu + Tsinghua University Fudan University University Pompeu Fabra Seoul National University 23
24 Summary We tackle the most challenging problems in Computer Science and related fields, affecting billions of users every day. Video Understanding Problem Large-Scale Challenges Kaggle Competition + CVPR 2017 Workshop 24
25 Thanks for your Attention!
YouTube-8M Video Classification
YouTube-8M Video Classification Alexandre Gauthier and Haiyu Lu Stanford University 450 Serra Mall Stanford, CA 94305 agau@stanford.edu hylu@stanford.edu Abstract Convolutional Neural Networks (CNNs) have
More informationLeveraging AI on the Cloud to transform your business. Florida Business Analytics Forum 2018 at University of South Florida
Leveraging AI on the Cloud to transform your business Florida Business Analytics Forum 2018 at University of South Florida 1 My (unusual) path to Google Neural networks at NOAA 2 DNNs solved image analysis
More informationInternet of things that video
Video recognition from a sentence Cees Snoek Intelligent Sensory Information Systems Lab University of Amsterdam The Netherlands Internet of things that video 45 billion cameras by 2022 [LDV Capital] 2
More informationWhat You Will Learn. What You Will Learn. How to Get Started with Wistia & 5 Ways It Generates More Leads. with Josh White
How to Get Started with Wistia & 5 Ways It Generates More Leads with Josh White What You Will Learn 1. Why Video Marketing 2. Importance to Businesses 3. Video Marketing Requirements 4. Video Platforms
More informationarxiv: v1 [cs.cv] 14 Jul 2017
Temporal Modeling Approaches for Large-scale Youtube-8M Video Understanding Fu Li, Chuang Gan, Xiao Liu, Yunlong Bian, Xiang Long, Yandong Li, Zhichao Li, Jie Zhou, Shilei Wen Baidu IDL & Tsinghua University
More informationUnstructured Data. CS102 Winter 2019
Winter 2019 Big Data Tools and Techniques Basic Data Manipulation and Analysis Performing well-defined computations or asking well-defined questions ( queries ) Data Mining Looking for patterns in data
More informationTowards Summarizing the Web of Entities
Towards Summarizing the Web of Entities contributors: August 15, 2012 Thomas Hofmann Director of Engineering Search Ads Quality Zurich, Google Switzerland thofmann@google.com Enrique Alfonseca Yasemin
More informationMulti-View 3D Object Detection Network for Autonomous Driving
Multi-View 3D Object Detection Network for Autonomous Driving Xiaozhi Chen, Huimin Ma, Ji Wan, Bo Li, Tian Xia CVPR 2017 (Spotlight) Presented By: Jason Ku Overview Motivation Dataset Network Architecture
More informationRECOMMENDATIONS HOW TO ATTRACT CLIENTS TO ROBOFOREX
RECOMMENDATIONS HOW TO ATTRACT CLIENTS TO ROBOFOREX Your success as a partner directly depends on the number of attracted clients and their trading activity. You can hardly influence clients trading activity,
More informationDeep Learning for Computer Vision with MATLAB By Jon Cherrie
Deep Learning for Computer Vision with MATLAB By Jon Cherrie 2015 The MathWorks, Inc. 1 Deep learning is getting a lot of attention "Dahl and his colleagues won $22,000 with a deeplearning system. 'We
More informationarxiv:submit/ [cs.cv] 16 Jun 2017
The Monkeytyping Solution to the YouTube-8M Video Understanding Challenge arxiv:submit/1922641 [cs.cv] 16 Jun 2017 He-Da Wang whd.thu@gmail.com Ji Wu Teng Zhang zhangteng1887@gmail.com wuji ee@mail.tsinghua.edu.cn
More informationThe Stanford/Technicolor/Fraunhofer HHI Video Semantic Indexing System
The Stanford/Technicolor/Fraunhofer HHI Video Semantic Indexing System Our first participation on the TRECVID workshop A. F. de Araujo 1, F. Silveira 2, H. Lakshman 3, J. Zepeda 2, A. Sheth 2, P. Perez
More informationSynscapes A photorealistic syntehtic dataset for street scene parsing Jonas Unger Department of Science and Technology Linköpings Universitet.
Synscapes A photorealistic syntehtic dataset for street scene parsing Jonas Unger Department of Science and Technology Linköpings Universitet 7D Labs VINNOVA https://7dlabs.com Photo-realistic image synthesis
More informationExploiting noisy web data for largescale visual recognition
Exploiting noisy web data for largescale visual recognition Lamberto Ballan University of Padova, Italy CVPRW WebVision - Jul 26, 2017 Datasets drive computer vision progress ImageNet Slide credit: O.
More informationApplication of Deep Learning Techniques in Satellite Telemetry Analysis.
Application of Deep Learning Techniques in Satellite Telemetry Analysis. Greg Adamski, Member of Technical Staff L3 Technologies Telemetry and RF Products Julian Spencer Jones, Spacecraft Engineer Telenor
More informationHow GPUs Power Comcast's X1 Voice Remote and Smart Video Analytics. Jan Neumann Comcast Labs DC May 10th, 2017
How GPUs Power Comcast's X1 Voice Remote and Smart Video Analytics Jan Neumann Comcast Labs DC May 10th, 2017 Comcast Applied Artificial Intelligence Lab Media & Video Analytics Smart TV Deep Learning
More informationA System for ecommerce Recommender Research with Context and Feedback
A System for ecommerce Recommender Research with Context and Feedback Sean Pfister RichRelevance http://code.richrelevance.com sean@richrelevance.com Agenda Introduction to {rr} and RecLab Review of context
More informationDeep Learning For Video Classification. Presented by Natalie Carlebach & Gil Sharon
Deep Learning For Video Classification Presented by Natalie Carlebach & Gil Sharon Overview Of Presentation Motivation Challenges of video classification Common datasets 4 different methods presented in
More informationPerson Action Recognition/Detection
Person Action Recognition/Detection Fabrício Ceschin Visão Computacional Prof. David Menotti Departamento de Informática - Universidade Federal do Paraná 1 In object recognition: is there a chair in the
More informationEncoder-Decoder Networks for Semantic Segmentation. Sachin Mehta
Encoder-Decoder Networks for Semantic Segmentation Sachin Mehta Outline > Overview of Semantic Segmentation > Encoder-Decoder Networks > Results What is Semantic Segmentation? Input: RGB Image Output:
More informationQuo Vadis, Action Recognition? A New Model and the Kinetics Dataset. By Joa õ Carreira and Andrew Zisserman Presenter: Zhisheng Huang 03/02/2018
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset By Joa õ Carreira and Andrew Zisserman Presenter: Zhisheng Huang 03/02/2018 Outline: Introduction Action classification architectures
More informationDepth from Stereo. Dominic Cheng February 7, 2018
Depth from Stereo Dominic Cheng February 7, 2018 Agenda 1. Introduction to stereo 2. Efficient Deep Learning for Stereo Matching (W. Luo, A. Schwing, and R. Urtasun. In CVPR 2016.) 3. Cascade Residual
More information2017 NACE Experience Conference July 16 19, 2017
NEXT LEVEL BRANDING From Business to Brand Presenter: Anja Winikka THE KNOT The Knot is #1 digital destination for couples planning their weddings with 11.5M monthly UVs (that s more than our next 4 competitors
More informationCS231N Section. Video Understanding 6/1/2018
CS231N Section Video Understanding 6/1/2018 Outline Background / Motivation / History Video Datasets Models Pre-deep learning CNN + RNN 3D convolution Two-stream What we ve seen in class so far... Image
More informationProject 3 Q&A. Jonathan Krause
Project 3 Q&A Jonathan Krause 1 Outline R-CNN Review Error metrics Code Overview Project 3 Report Project 3 Presentations 2 Outline R-CNN Review Error metrics Code Overview Project 3 Report Project 3 Presentations
More informationAudioSet: Real-world Audio Event Classification
AudioSet: Real-world Audio Event Classification g.co/audioset Rif A. Saurous, Shawn Hershey, Dan Ellis, Aren Jansen and the Google Sound Understanding Team 2017-10-20 Outline The Early Years: Weakly-Supervised
More informationSegmentation. Bottom up Segmentation Semantic Segmentation
Segmentation Bottom up Segmentation Semantic Segmentation Semantic Labeling of Street Scenes Ground Truth Labels 11 classes, almost all occur simultaneously, large changes in viewpoint, scale sky, road,
More informationCS 523: Multimedia Systems
CS 523: Multimedia Systems Angus Forbes creativecoding.evl.uic.edu/courses/cs523 Today - Convolutional Neural Networks - Work on Project 1 http://playground.tensorflow.org/ Convolutional Neural Networks
More informationDemystifying Deep Learning
Demystifying Deep Learning Let the computers do the hard work Jérémy Huard 2015 The MathWorks, Inc. 1 2 Why MATLAB for Deep Learning? MATLAB is Productive MATLAB is Fast MATLAB Integrates with Open Source
More informationHide-and-Seek: Forcing a network to be Meticulous for Weakly-supervised Object and Action Localization
Hide-and-Seek: Forcing a network to be Meticulous for Weakly-supervised Object and Action Localization Krishna Kumar Singh and Yong Jae Lee University of California, Davis ---- Paper Presentation Yixian
More informationJoint Inference in Image Databases via Dense Correspondence. Michael Rubinstein MIT CSAIL (while interning at Microsoft Research)
Joint Inference in Image Databases via Dense Correspondence Michael Rubinstein MIT CSAIL (while interning at Microsoft Research) My work Throughout the year (and my PhD thesis): Temporal Video Analysis
More informationCAP 6412 Advanced Computer Vision
CAP 6412 Advanced Computer Vision http://www.cs.ucf.edu/~bgong/cap6412.html Boqing Gong April 21st, 2016 Today Administrivia Free parameters in an approach, model, or algorithm? Egocentric videos by Aisha
More informationDeep Character-Level Click-Through Rate Prediction for Sponsored Search
Deep Character-Level Click-Through Rate Prediction for Sponsored Search Bora Edizel - Phd Student UPF Amin Mantrach - Criteo Research Xiao Bai - Oath This work was done at Yahoo and will be presented as
More information2 The IBM Data Governance Unified Process
2 The IBM Data Governance Unified Process The benefits of a commitment to a comprehensive enterprise Data Governance initiative are many and varied, and so are the challenges to achieving strong Data Governance.
More informationExclusive Leads for Attorneys WHY PAY PER CLICK?
Exclusive Leads for Attorneys Lead Generation Program WHY PAY PER CLICK? BusinessCreator, Inc. 855-943-8736 marketing@forlawfirmsonly.com BusinessCreatorPlus.com ForLawFirmsOnly.com Power Practice Builder
More informationGet More Out of Hitting Record IT S EASY TO CREATE EXCEPTIONAL VIDEO CONTENT WITH MEDIASITE JOIN
Get More Out of Hitting Record IT S EASY TO CREATE EXCEPTIONAL VIDEO CONTENT WITH MEDIASITE JOIN Better Video Starts With Better Capture Too often, great ideas and important details are lost when the video
More informationFunctionalities & Applications. D. M. Gavrila (UvA) and E. Jansen (TNO)
Functionalities & Applications D. M. Gavrila (UvA) and E. Jansen (TNO) Algorithms, Functionalities and Applications Algorithms (Methods) Functionalities (Application building blocks) Applications (Systems)
More informationClass 9 Action Recognition
Class 9 Action Recognition Liangliang Cao, April 4, 2013 EECS 6890 Topics in Information Processing Spring 2013, Columbia University http://rogerioferis.com/visualrecognitionandsearch Visual Recognition
More informationStep 1: Open browser to navigate to the data science challenge home page
Step 1: Open browser to navigate to the data science challenge home page https://datascience.ey.com/ Step 2: Logging in You will need to create an account if you are a new user. Click the sign up button
More informationIndoor Object Recognition of 3D Kinect Dataset with RNNs
Indoor Object Recognition of 3D Kinect Dataset with RNNs Thiraphat Charoensripongsa, Yue Chen, Brian Cheng 1. Introduction Recent work at Stanford in the area of scene understanding has involved using
More information3D Shape Analysis with Multi-view Convolutional Networks. Evangelos Kalogerakis
3D Shape Analysis with Multi-view Convolutional Networks Evangelos Kalogerakis 3D model repositories [3D Warehouse - video] 3D geometry acquisition [KinectFusion - video] 3D shapes come in various flavors
More informationExploring World s Interest in Paralympics through Twitter
Exploring World s Interest in Paralympics through Twitter Venkata Sravya Kalla, Thanaa Ghanem Information and Computer Science Department Metropolitan State University St. Paul, MN, 55106 cu9426bs@metrostate.edu,
More informationIntelligent Edge Computing and ML-based Traffic Classifier. Kwihoon Kim, Minsuk Kim (ETRI) April 25.
Intelligent Edge Computing and ML-based Traffic Classifier Kwihoon Kim, Minsuk Kim (ETRI) (kwihooi@etri.re.kr, mskim16@etri.re.kr) April 25. 2018 ITU Workshop on Impact of AI on ICT Infrastructures Cian,
More informationObject Detection by 3D Aspectlets and Occlusion Reasoning
Object Detection by 3D Aspectlets and Occlusion Reasoning Yu Xiang University of Michigan Silvio Savarese Stanford University In the 4th International IEEE Workshop on 3D Representation and Recognition
More informationLearning Semantic Video Captioning using Data Generated with Grand Theft Auto
A dark car is turning left on an exit Learning Semantic Video Captioning using Data Generated with Grand Theft Auto Alex Polis Polichroniadis Data Scientist, MSc Kolia Sadeghi Applied Mathematician, PhD
More informationTizen apps with. Context Awareness, powered by AI. by Shashwat Pradhan, CEO Emberify
Tizen apps with 1 Context Awareness, powered by AI by Shashwat Pradhan, CEO Emberify Introduction Context refers to information that characterizes a situation, between: Apps People Surrounding environment
More informationDefinition, Detection, and Evaluation of Meeting Events in Airport Surveillance Videos
Definition, Detection, and Evaluation of Meeting Events in Airport Surveillance Videos Sung Chun Lee, Chang Huang, and Ram Nevatia University of Southern California, Los Angeles, CA 90089, USA sungchun@usc.edu,
More informationLecture 7: Semantic Segmentation
Semantic Segmentation CSED703R: Deep Learning for Visual Recognition (207F) Segmenting images based on its semantic notion Lecture 7: Semantic Segmentation Bohyung Han Computer Vision Lab. bhhanpostech.ac.kr
More informationPointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space Sikai Zhong February 14, 2018 COMPUTER SCIENCE Table of contents 1. PointNet 2. PointNet++ 3. Experiments 1 PointNet Property
More informationProgramming Projects
Programming Projects Benjamin Roth, Nina Poerner, Anne Beyer Centrum für Informations- und Sprachverarbeitung Ludwig-Maximilian-Universität München beroth@cis.uni-muenchen.de Benjamin Roth, Nina Poerner,
More informationCONSUMERLAB. Liberation from location. Consumers developing place-agnostic internet habits
CONSUMERLAB Liberation from location Consumers developing place-agnostic internet habits An Ericsson Consumer Insight Summary Report October 2014 contents THE NEED TO KNOW 3 CONVERGING HABITS 4 FREEDOM
More informationOverview. Data-mining. Commercial & Scientific Applications. Ongoing Research Activities. From Research to Technology Transfer
Data Mining George Karypis Department of Computer Science Digital Technology Center University of Minnesota, Minneapolis, USA. http://www.cs.umn.edu/~karypis karypis@cs.umn.edu Overview Data-mining What
More informationBuilding a Restaurant Menu Presentation Database and Visualization
Building a Restaurant Menu Presentation Database and Visualization Yeong Hyeon Gu, Seong Joon Yoo*, Dongil Han, Sung Wook Baek, Byung-Joo Shin, and Yun Hwan Kim Abstract For restaurants successful advancement
More informationBest of SharePoint Sites and Communities
Best of SharePoint 2010 Sites and Communities Agenda Overview and SharePoint 2010 Basics SharePoint Foundation Sites Communities Business Needs IT Needs Microsoft SharePoint 2010 The business collaboration
More informationKS Blogs Tutorial Wikipedia definition of a blog : Some KS Blog definitions: Recommendation:
KS Blogs Tutorial Wikipedia definition of a blog : A blog (a portmanteau of web log) is a website where entries are written in chronological order and commonly displayed in reverse chronological order.
More informationPerson Re-identification for Improved Multi-person Multi-camera Tracking by Continuous Entity Association
Person Re-identification for Improved Multi-person Multi-camera Tracking by Continuous Entity Association Neeti Narayan, Nishant Sankaran, Devansh Arpit, Karthik Dantu, Srirangaraj Setlur, Venu Govindaraju
More informationInference Optimization Using TensorRT with Use Cases. Jack Han / 한재근 Solutions Architect NVIDIA
Inference Optimization Using TensorRT with Use Cases Jack Han / 한재근 Solutions Architect NVIDIA Search Image NLP Maps TensorRT 4 Adoption Use Cases Speech Video AI Inference is exploding 1 Billion Videos
More informationInception and Residual Networks. Hantao Zhang. Deep Learning with Python.
Inception and Residual Networks Hantao Zhang Deep Learning with Python https://en.wikipedia.org/wiki/residual_neural_network Deep Neural Network Progress from Large Scale Visual Recognition Challenge (ILSVRC)
More informationATA DRIVEN GLOBAL VISION CLOUD PLATFORM STRATEG N POWERFUL RELEVANT PERFORMANCE SOLUTION CLO IRTUAL BIG DATA SOLUTION ROI FLEXIBLE DATA DRIVEN V
ATA DRIVEN GLOBAL VISION CLOUD PLATFORM STRATEG N POWERFUL RELEVANT PERFORMANCE SOLUTION CLO IRTUAL BIG DATA SOLUTION ROI FLEXIBLE DATA DRIVEN V WHITE PAPER Create the Data Center of the Future Accelerate
More informationPersonalizing Netflix with Streaming datasets
Personalizing Netflix with Streaming datasets Shriya Arora Senior Data Engineer Personalization Analytics @shriyarora What is this talk about? Helping you decide if a streaming pipeline fits your ETL problem
More informationGoogle GSuite Intro Demo of GSuite and GCP integration
Google GSuite Intro Demo of GSuite and GCP integration May 2017 Sara Djelassi - Sales Steve Mansfield - PSO 7 Cloud products with 1 billion users ML is core to differentiating Google services Search Search
More informationAutoCalib: Automatic Calibration of Traffic Cameras at Scale
AutoCalib: Automatic of Traffic Cameras at Scale Romil Bhardwaj, Gopi Krishna Tummala*, Ganesan Ramalingam, Ramachandran Ramjee, Prasun Sinha* Microsoft Research, *The Ohio State University Number of Cameras
More informationVISION FOR AUTOMOTIVE DRIVING
VISION FOR AUTOMOTIVE DRIVING French Japanese Workshop on Deep Learning & AI, Paris, October 25th, 2017 Quoc Cuong PHAM, PhD Vision and Content Engineering Lab AI & MACHINE LEARNING FOR ADAS AND SELF-DRIVING
More informationOverview of the 2013 ALTA Shared Task
Overview of the 2013 ALTA Shared Task Diego Molla Department of Computing Macquarie University Sydney, NSW 2109 diego.molla-aliod@mq.edu.au Abstract The 2013 ALTA shared task was the fourth in the ALTA
More informationNIS Directive : Call for Proposals
National Cyber Security Centre, in Collaboration with the Research Institute in Trustworthy Inter-connected Cyber-physical Systems (RITICS) Summary NIS Directive : Call for Proposals Closing date: Friday
More informationIntroduction to Deep Learning in Signal Processing & Communications with MATLAB
Introduction to Deep Learning in Signal Processing & Communications with MATLAB Dr. Amod Anandkumar Pallavi Kar Application Engineering Group, Mathworks India 2019 The MathWorks, Inc. 1 Different Types
More information2015 The MathWorks, Inc. 1
2015 The MathWorks, Inc. 1 개발에서구현까지 MATLAB 환경에서의딥러닝 김종남 Application Engineer 2015 The MathWorks, Inc. 2 3 Why MATLAB for Deep Learning? MATLAB is Productive MATLAB is Fast MATLAB Integrates with Open Source
More informationCurriculum Map: Digital Communications MASH Communications Department
Curriculum Map: Digital Communications MASH Communications Department Course Description: This semester long course is designed to introduce students to techniques required to communicate in a 21 st century
More informationSEARCHMETRICS WHITEPAPER RANKING FACTORS Targeted Analysis for more Success on Google and in your Online Market
2018 SEARCHMETRICS WHITEPAPER RANKING FACTORS 2018 Targeted for more Success on Google and in your Online Market Table of Contents Introduction: Why ranking factors for niches?... 3 Methodology: Which
More informationAnimation tools. Using Go!Animate
Animation tools Visual displays are often the most effective way to get a message across, particularly if there are numbers involved. As well as charting tools for numeric comparisons and predictions,
More informationDeep Learning for Recommender Systems
join at Slido.com with #bigdata2018 Deep Learning for Recommender Systems Oliver Gindele @tinyoli oliver.gindele@datatonic.com Big Data Conference Vilnius 28.11.2018 Who is Oliver? + Head of Machine Learning
More informationEnhancing applications with Cognitive APIs IBM Corporation
Enhancing applications with Cognitive APIs After you complete this section, you should understand: The Watson Developer Cloud offerings and APIs The benefits of commonly used Cognitive services 2 Watson
More informationTri-modal Human Body Segmentation
Tri-modal Human Body Segmentation Master of Science Thesis Cristina Palmero Cantariño Advisor: Sergio Escalera Guerrero February 6, 2014 Outline 1 Introduction 2 Tri-modal dataset 3 Proposed baseline 4
More informationDeconvolution Networks
Deconvolution Networks Johan Brynolfsson Mathematical Statistics Centre for Mathematical Sciences Lund University December 6th 2016 1 / 27 Deconvolution Neural Networks 2 / 27 Image Deconvolution True
More informationDeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution and Fully Connected CRFs
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution and Fully Connected CRFs Zhipeng Yan, Moyuan Huang, Hao Jiang 5/1/2017 1 Outline Background semantic segmentation Objective,
More informationWhat are we trying to achieve? Why are we doing this? What do we learn from past history? What will we talk about today?
Introduction What are we trying to achieve? Why are we doing this? What do we learn from past history? What will we talk about today? What are we trying to achieve? Example from Scott Satkin 3D interpretation
More informationFoundations for Summarizing and Learning Latent Structure in Video
Foundations for Summarizing and Learning Latent Structure in Video Presenter: Kevin Pitstick, MTS Engineer PI: Ed Morris, MTS Senior Engineer Copyright 2017 Carnegie Mellon University. All Rights Reserved.
More informationColumbia University High-Level Feature Detection: Parts-based Concept Detectors
TRECVID 2005 Workshop Columbia University High-Level Feature Detection: Parts-based Concept Detectors Dong-Qing Zhang, Shih-Fu Chang, Winston Hsu, Lexin Xie, Eric Zavesky Digital Video and Multimedia Lab
More informationLinear combinations of simple classifiers for the PASCAL challenge
Linear combinations of simple classifiers for the PASCAL challenge Nik A. Melchior and David Lee 16 721 Advanced Perception The Robotics Institute Carnegie Mellon University Email: melchior@cmu.edu, dlee1@andrew.cmu.edu
More informationOffering Access to Personalized Interactive Video
Offering Access to Personalized Interactive Video 1 Offering Access to Personalized Interactive Video Giorgos Andreou, Phivos Mylonas, Manolis Wallace and Stefanos Kollias Image, Video and Multimedia Systems
More informationRUSSIA AUTOMOTIVE INDUSTRY TRENDS
RUSSIA AUTOMOTIVE INDUSTRY TRENDS 2017 TABLE OF CONTENTS CATEGORY TRENDS & KEY AUCTION METRICS BRAND LEADERBOARD ON GOOGLE SEARCH AUTOMOTIVE TRENDS ON YOUTUBE I Key 2017 highlights 01 Automotive queries
More informationETISEO, performance evaluation for video surveillance systems
ETISEO, performance evaluation for video surveillance systems A. T. Nghiem, F. Bremond, M. Thonnat, V. Valentin Project Orion, INRIA - Sophia Antipolis France Abstract This paper presents the results of
More informationContexts and 3D Scenes
Contexts and 3D Scenes Computer Vision Jia-Bin Huang, Virginia Tech Many slides from D. Hoiem Administrative stuffs Final project presentation Dec 1 st 3:30 PM 4:45 PM Goodwin Hall Atrium Grading Three
More informationLSTM and its variants for visual recognition. Xiaodan Liang Sun Yat-sen University
LSTM and its variants for visual recognition Xiaodan Liang xdliang328@gmail.com Sun Yat-sen University Outline Context Modelling with CNN LSTM and its Variants LSTM Architecture Variants Application in
More informationAutomatic people tagging for expertise profiling in the enterprise
Automatic people tagging for expertise profiling in the enterprise Pavel Serdyukov * (Yandex, Moscow, Russia) Mike Taylor, Vishwa Vinay, Matthew Richardson, Ryen White (Microsoft Research, Cambridge /
More informationDeep Incremental Scene Understanding. Federico Tombari & Christian Rupprecht Technical University of Munich, Germany
Deep Incremental Scene Understanding Federico Tombari & Christian Rupprecht Technical University of Munich, Germany C. Couprie et al. "Toward Real-time Indoor Semantic Segmentation Using Depth Information"
More informationEND-TO-END CHINESE TEXT RECOGNITION
END-TO-END CHINESE TEXT RECOGNITION Jie Hu 1, Tszhang Guo 1, Ji Cao 2, Changshui Zhang 1 1 Department of Automation, Tsinghua University 2 Beijing SinoVoice Technology November 15, 2017 Presentation at
More informationLive Streaming to Internal, Remote and External Locations. A PTZOptics Live Presentation
Live Streaming to Internal, Remote and External Locations A PTZOptics Live Presentation What you will find in this guide 1. 2. 3. 4. 5. 6. 7. High Level Networking Concepts Working with Internal, Remote
More informationIs Bigger CNN Better? Samer Hijazi on behalf of IPG CTO Group Embedded Neural Networks Summit (enns2016) San Jose Feb. 9th
Is Bigger CNN Better? Samer Hijazi on behalf of IPG CTO Group Embedded Neural Networks Summit (enns2016) San Jose Feb. 9th Today s Story Why does CNN matter to the embedded world? How to enable CNN in
More informationZOOM Video Conferencing: Quick Start Guide
ZOOM Video Conferencing: Quick Start Guide Welcome to Zoom at James Cook University (JCU), a video conferencing system designed to enhance your communication and collaboration with colleagues, students
More informationStreaming videos. Problem statement for Online Qualification Round, Hash Code 2017
Streaming videos Problem statement for Online Qualification Round, Hash Code 2017 Introduction Have you ever wondered what happens behind the scenes when you watch a YouTube video? As more and more people
More informationROB 537: Learning-Based Control
ROB 537: Learning-Based Control Week 6, Lecture 1 Deep Learning (based on lectures by Fuxin Li, CS 519: Deep Learning) Announcements: HW 3 Due TODAY Midterm Exam on 11/6 Reading: Survey paper on Deep Learning
More informationAWS DeepLens Workshop: Building a Computer Vision App
AWS DeepLens Workshop: Building a Computer Vision App Jyothi Nookula - Senior Product Manager, Amazon Web Services May 23 rd 2018 AWS DeepLens is not a video camera I t s t h e w o r l d s f i r s t d
More informationVolume 6, Issue 12, December 2018 International Journal of Advance Research in Computer Science and Management Studies
ISSN: 2321-7782 (Online) e-isjn: A4372-3114 Impact Factor: 7.327 Volume 6, Issue 12, December 2018 International Journal of Advance Research in Computer Science and Management Studies Research Article
More informationA Study on Multi-resolution Screen based Conference Broadcasting Technology
2 : (Young-ae Kim et al.: A Study on Multi-resolution Screen based Conference Broadcasting Technology) (Special Paper) 23 2, 2018 3 (JBE Vol. 23, No. 2, March 2018) https://doi.org/10.5909/jbe.2018.23.2.253
More informationBPMR Mission: Korea. Industry to Industry Dialogue on Emissions Trading and Market Readiness
BPMR Mission: Korea Industry to Industry Dialogue on Emissions Trading and Market Readiness I. Introduction March 23-24 Hoam Faculty House in Seoul National University Seoul, Republic of Korea (website
More informationClass 5: Attributes and Semantic Features
Class 5: Attributes and Semantic Features Rogerio Feris, Feb 21, 2013 EECS 6890 Topics in Information Processing Spring 2013, Columbia University http://rogerioferis.com/visualrecognitionandsearch Project
More informationHow to Build Optimized ML Applications with Arm Software
How to Build Optimized ML Applications with Arm Software Arm Technical Symposia 2018 Arm K.K. Senior FAE Ryuji Tanaka Overview Today we will talk about applied machine learning (ML) on Arm. My aim for
More informationHuman Pose Estimation with Deep Learning. Wei Yang
Human Pose Estimation with Deep Learning Wei Yang Applications Understand Activities Family Robots American Heist (2014) - The Bank Robbery Scene 2 What do we need to know to recognize a crime scene? 3
More informationHierarchical Video Frame Sequence Representation with Deep Convolutional Graph Network
Hierarchical Video Frame Sequence Representation with Deep Convolutional Graph Network Feng Mao [0000 0001 6171 3168], Xiang Wu [0000 0003 2698 2156], Hui Xue, and Rong Zhang Alibaba Group, Hangzhou, China
More information