ISBN:
9780128092804
Language:
English
Pages:
1 Online-Ressource (440 pages)
Parallel Title:
Print version Murino, Vittorio Group and Crowd Behavior for Computer Vision
DDC:
006.37
Keywords:
Computer vision
;
Computer vision
;
Electronic books
Abstract:
Front Cover -- Group and Crowd Behavior for Computer Vision -- Copyright -- Contents -- About the Editors -- 1 The Group and Crowd Analysis Interdisciplinary Challenge -- 1.1 The Study of Groups and Crowds -- 1.2 Scope of the Book -- 1.3 Summary of Important Points -- References -- Part 1 Features and Representations -- 2 Social Interaction in Temporary Gatherings -- 2.1 Introduction: Group and Crowd Behavior in Context -- 2.2 Social Interaction: A Typology and Some De nitions -- 2.2.1 Unfocused Interaction -- 2.2.2 Common-Focused Interaction -- 2.2.3 Jointly-Focused Interaction -- 2.3 Temporary Gatherings: A Taxonomy and Some Examples -- 2.3.1 Small Gatherings - Semi/Private Encounters and Group Life -- 2.3.2 Medium Gatherings - Semi/Public Occasions and Community Life -- 2.3.3 Large Gatherings - Public Events and Collective Life -- 2.4 Conclusion: Microsociology Applied to Computer Vision -- 2.5 Further Reading -- References -- 3 Group Detection and Tracking Using Sociological Features -- 3.1 Introduction -- 3.2 State-of-the-Art -- 3.3 Sociological Features -- 3.3.1 Low-Level Features -- 3.3.1.1 Person Detection -- 3.3.1.2 Person Velocity and Direction -- 3.3.1.3 Head & Body Orientation -- 3.3.2 High-Level Features -- 3.3.2.1 3D Subjective View Frustum -- 3.3.2.2 Transactional Segment-Based Frustum -- 3.3.2.3 External Factors -- 3.4 Detection Models -- 3.4.1 Game-Theoretic Conversational Grouping Model -- 3.4.2 The Dirichlet Process Mixture Model -- 3.5 Group Tracking -- 3.5.1 DPF for Group Tracking -- Individual Proposal p(Xt+1|X0:t,y0:t+1) -- Joint Observation Distribution p(yt|Xt ,Tt ) -- Joint Individual Distribution p( Xt+1|X0:t,Tt ) -- Joint Group Proposal p( Tt+1|X0:t+1,Tt ) -- 3.6 Experiments -- 3.6.1 Results of Group Detection -- 3.6.1.1 Datasets -- 3.6.1.2 Evaluation Metrics -- 3.6.1.3 Comparing Methods
Abstract:
3.6.1.4 Performance Evaluation -- 3.6.2 Results of Group Tracking -- 3.6.2.1 Datasets -- 3.6.2.2 Evaluation Metrics -- 3.6.2.3 Comparing Methods -- 3.6.2.4 Performance Analysis -- 3.7 Discussion -- 3.8 Conclusions -- References -- 4 Exploring Multitask and Transfer Learning Algorithms for Head Pose Estimation in Dynamic Multiview Scenarios -- 4.1 Introduction -- 4.2 Related Work -- 4.2.1 Head Pose Estimation from Low-Resolution Images -- 4.2.2 Transfer Learning -- 4.2.3 Multitask Learning -- 4.3 TL and MTL for Multiview Head Pose Estimation -- 4.3.1 Preprocessing -- 4.3.2 Transfer Learning for HPE -- 4.3.2.1 Head-Pan Classi cation Under Varying Head-Tilt -- Experimental Results -- 4.3.2.2 Head-Pan Classi cation Under Target Motion -- Experimental Results -- 4.3.3 Multitask Learning for HPE -- 4.3.3.1 FEGA-MTL -- Experimental Results -- 4.4 Conclusions -- References -- 5 The Analysis of High Density Crowds in Videos -- 5.1 Introduction -- 5.2 Literature Review -- 5.2.1 Crowd Motion Modeling and Segmentation -- 5.2.2 Estimating Density of People in a Crowded Scene -- 5.2.3 Crowd Event Modeling and Recognition -- 5.2.4 Detecting and Tracking in a Crowded Scene -- 5.3 Data-Driven Crowd Analysis in Videos -- 5.3.1 Off-Line Analysis of Crowd Video Database -- 5.3.1.1 Low-Level Representation -- 5.3.1.2 Mid-Level Representation -- 5.3.2 Matching -- 5.3.2.1 Global Crowded Scene Matching -- 5.3.2.2 Local Crowd Patch Matching -- 5.3.3 Transferring Learned Crowd Behaviors -- 5.3.4 Experiments and Results -- 5.4 Density-Aware Person Detection and Tracking in Crowds -- 5.4.1 Crowd Model -- 5.4.1.1 Tracking Detections -- 5.4.2 Evaluation -- 5.4.2.1 Tracking -- 5.5 CrowdNet: Learning a Representation for High Density Crowds in Videos -- 5.5.1 Introduction -- 5.5.2 Overview of the Approach -- 5.5.3 Crowd Patch Mining in Videos -- 5.5.4 Tracking
Abstract:
5.5.5 Learning a Representation for High Density Crowds -- 5.5.6 Evaluation -- 5.6 Conclusions and Directions for Future Research -- References -- 6 Tracking Millions of Humans in Crowded Spaces -- 6.1 Introduction -- 6.2 Related Work -- 6.3 System Overview -- 6.4 Human Detection in 3D -- 6.4.1 Method -- 6.4.2 Evaluation -- 6.5 Tracklet Generation -- 6.6 Tracklet Association -- 6.6.1 Social Af nity Map - SAM -- 6.6.2 The SAM Feature -- 6.6.3 Tracklet Association Method -- 6.6.4 Optimization -- 6.6.5 Coarse-to-Fine Data Association -- 6.7 Experiments -- 6.7.1 Large-Scale Evaluation -- 6.7.2 OD Forecasting -- 6.8 Conclusions -- References -- 7 Subject-Centric Group Feature for Person Reidenti cation -- 7.1 Introduction -- 7.2 Related Works -- 7.3 Methodology -- 7.3.1 Group Extraction -- 7.3.2 Person-Group Feature -- 7.3.2.1 In-Group Position Signature -- 7.3.2.2 Metric of Person-Group Feature -- 7.3.3 Person Reidenti cation with Person-Group Feature -- 7.4 Results -- 7.4.1 Features Evaluation -- 7.4.1.1 Group Extraction Evaluation -- 7.4.1.2 Group Features Evaluation -- 7.4.2 Comparison with Baseline Approaches -- 7.4.3 Comparison with Group-Based Approaches -- 7.5 Conclusion -- Acknowledgments -- References -- Part 2 Group and Crowd Behavior Modeling -- 8 From Groups to Leaders and Back -- 8.1 Introduction -- 8.2 Modeling and Observing Groups and Their Leaders in Literature -- 8.2.1 Sociological Perspective -- 8.2.2 Computational Approaches -- 8.3 Technical Preliminaries and Structured Output Prediction -- 8.3.1 Problem Statement -- 8.3.2 Stochastic Optimization -- 8.4 The Tools of the Trade in Social and Structured Crowd Analysis -- 8.4.1 Socially Constrained Structural Learning for Groups Detection in Crowd -- 8.4.1.1 Task Formulation -- 8.4.1.2 SSVM Adaptation to Group Detection -- Inference and Max Oracle -- Loss Function
Abstract:
8.4.2 Learning to Identify Group Leaders in Crowd -- 8.4.2.1 Task Formulation -- 8.4.2.2 SSVM Adaptation to Leader Identi cation -- Inference and Max Oracle -- Loss Function -- 8.5 Results on Visual Localization of Groups and Leaders -- 8.6 The Predictive Power of Leaders in Social Groups -- 8.6.1 Experimental Settings -- 8.6.2 Leader Centrality in Feature Space -- 8.6.2.1 Group Recovery Guarantees -- 8.6.2.2 Validation and Results -- 8.7 Conclusion -- References -- 9 Learning to Predict Human Behavior in Crowded Scenes -- 9.1 Introduction -- 9.2 Related Work -- 9.2.1 Human-Human Interactions -- 9.2.2 Activity Forecasting -- 9.2.3 RNN Models for Sequence Prediction -- 9.3 Forecasting with Social Forces Model -- 9.3.1 Basic Theory -- 9.3.2 Modeling Social Sensitivity -- 9.3.2.1 Social Sensitivity Feature -- 9.3.2.2 Training -- 9.3.2.3 Testing -- 9.3.3 Forecasting with Social Sensitivity -- 9.4 Forecasting with Recurrent Neural Network -- 9.4.1 Social LSTM -- 9.4.1.1 Social Pooling of Hidden States -- 9.4.1.2 Position Estimation -- 9.4.1.3 Occupancy Map Pooling -- 9.4.1.4 Inference for Path Prediction -- 9.4.2 Implementation Details -- 9.5 Experiments -- 9.5.1 Analyzing the Predicted Paths -- 9.5.2 Discussions and Limitations -- 9.6 Conclusions -- References -- 10 Deep Learning for Scene-Independent Crowd Analysis -- 10.1 Introduction -- 10.2 Large Scale Crowd Datasets -- 10.2.1 Shanghai World Expo'10 Crowd Dataset -- 10.2.1.1 Data Collection -- 10.2.1.2 Annotation -- 10.2.2 WWW Crowd Dataset -- 10.2.2.1 Crowd Video Construction -- Collecting Keywords -- Collecting Crowd Videos -- 10.2.2.2 Crowd Attribute Annotation -- Collecting Crowd Attributes from Web Tags -- Crowd Attribute Annotation -- 10.2.3 User Study on Crowd Attribute -- 10.3 Crowd Counting and Density Estimation -- 10.3.1 Method -- 10.3.1.1 Normalized Crowd Density Map for Training
Abstract:
10.3.1.2 Crowd CNN Model -- 10.3.2 Nonparametric Fine-Tuning Method for Target Scene -- 10.3.2.1 Candidate Fine-Tuning Scene Retrieval -- 10.3.2.2 Local Patch Retrieval -- 10.3.2.3 Experimental Results -- 10.4 Attributes for Crowded Scene Understanding -- 10.4.1 Related Work -- 10.4.2 Slicing Convolutional Neural Network -- 10.4.2.1 Semantic Selectiveness of Feature Maps -- 10.4.2.2 Feature Map Pruning -- Af nity Score -- Conspicuous Score -- 10.4.2.3 Semantic Temporal Slices -- 10.4.3 S-CNN Deep Architecture -- 10.4.3.1 Single Branch of S-CNN Model -- S-CNN-xy Branch -- S-CNN-xt/-yt Branch -- 10.4.3.2 Combined S-CNN Model -- 10.4.4 Experiments -- 10.4.4.1 Experimental Setting -- Dataset -- Evaluation Metrics -- Model Pre-Training -- 10.4.4.2 Ablation Study of S-CNN -- Level of Semantics and Temporal Range -- Pruning of Features -- Single Branch Model vs. Combined Model -- 10.4.4.3 Comparison with State-of-the-Art Methods -- Quantitative Evaluation -- Qualitative Evaluation -- 10.5 Conclusion -- References -- 11 Physics-Inspired Models for Detecting Abnormal Behaviors in Crowded Scenes -- 11.1 Introduction -- 11.2 Crowd Anomaly Detection: A General Review -- 11.3 Physics-Inspired Crowd Models -- 11.3.1 Social Force Models -- 11.3.2 Flow Field Models -- 11.3.3 Crowd Energy Models -- 11.3.4 Substantial Derivative -- 11.4 Violence Detection -- 11.4.1 The Substantial Derivative Model -- 11.4.1.1 Substantial Derivative in Fluid Mechanics -- 11.4.1.2 Modeling Pedestrian Motion Dynamics -- 11.4.1.3 Estimation of Local and Convective Forces from Videos -- 11.5 Experimental Results -- 11.5.1 Datasets -- 11.5.2 Effect of Sampled Patches -- 11.5.3 Comparison to State-of-the-Art -- 11.6 Conclusions -- References -- 12 Activity Forecasting -- 12.1 Introduction -- 12.2 Overview -- 12.3 Activity Forecasting as Optimal Control
Abstract:
12.3.1 Toward Decision-Theoretic Models
URL:
Volltext
(lizenzpflichtig)
Permalink
