Distributed Video Sensor Networks

Distributed Video Sensor Networks\n	Preface\n	Contents\n	Introduction\nDistributed Video Sensor Networks and Research Challenges\n	Report on NSF/ARO/ONR Workshop on Distributed Camera Networks: Research Challenges and Future Directions\n		Introduction\n		Workshop Recommendations\n			A. Video Processing and Video Understanding\n				Calibration Risk and Payoffs\n				Nuisance Factors Risks and Payoffs\n			B. Simulation, Graphics, Cognition and Video Networks\n			C. Wireless Video Sensor Networks, Communications and Control\n			D. Distributed Embedded Cameras and Real Time Video Analysis\n			E. Applications\n			F. Educational Opportunities and Curriculum Development\n		Suggested Major Research Topics\n			Topic 1: Robust Scalable Video Networks for Wide Area Analysis\n				Disciplines Involved\n				Research Concentration\n			Topic 2: Active, Distributed and Communication Aware Video Sensor Networks\n				Disciplines Involved\n				Research Concentration\n			Topic 3: Large-scale Heterogeneous Sensor Networks for Wide Area Analysis\n				Disciplines Involved\n				Research Concentration\n		List of Attendees in Alphabetical Order\n		Groups and Group Leaders\n		Talks with Titles and Presenters\nVideo Processing and Understanding\n	Motion Analysis: Past, Present and Future\n		Introduction to Motion: An Early History\n		Motion: Highlights from Philosophy, Psychology and Neurobiology\n		Motion in Computer Vision: The Beginnings\n		Optical Flow-Based Motion Detection\n		Human Actions and Activities\n		Motion: Future\n		References\n	Projective Joint Invariants for Matching Curves in Camera Networks\n		Introduction\n			Related Work\n			Our Approach\n		Problem Formulation and Preliminaries\n		Joint-Invariant Signatures\n		Toward Local Signatures\n			Slices and Sections of Signature Manifold\n			Correspondence and Equivalence from Matching Sections\n			Picking Pivot Points\n		Matching Performance\n		Discussion\n		References\n	Multiple-View Object Recognition in Smart Camera Networks\n		Introduction\n			Contributions\n		Encoding Multiple-View Features via Sparse Representation\n			Random Projections\n			Enforcing Nonnegativity in l1-Minimization\n			Estimation of Joint Sparse Signals\n		System Implementation\n		Experiment\n		Conclusion and Discussion\n		References\n	A Comparison of Techniques for Camera Selection and Hand-Off in a Video Network\n		Introduction\n		Related Work and Contributions\n			Comparison for Existing Works\n			Our Contributions\n		Theoretical Comparison\n			Descriptions of the Key Ideas of Selected Approaches\n				The Utility-Based Game Theoretic Approach\n				The Co-occurrence to Occurrence Ratio (COR) Approach\n				The Constraint Satisfaction Problem (CSP) Approach\n				The Fuzzy-based Approach\n		Experimental Results\n			Data\n			Tracking\n			Parameters\n			Experimental Results and Analysis\n		Conclusions and Future Work\n		References\n	Distributed Sensing and Processing for Multi-Camera Networks\n		Introduction\n			Robust Statistical Inference\n			Computationally Efficient and Distributed Algorithms\n			Opportunistic and Parsimonious Sensing\n		Statistical Inference for Tracking\n			Homography\n			Detection\n			Multi-View Fusion and Tracking\n		Efficient Particle Filtering\n			Particle Filtering: A Brief Overview\n			Metropolis Hastings Algorithm\n			Particle Filtering with IMHA-Based Resampling\n			Experimental Results\n		Compressive Sensing\n			Compressive Sensing\n			Compressive Background Subtraction\n			Multi-View Ground-Plane Tracking\n		Conclusions and Future Directions\n			Distributed Bayesian Inference\n			Manifold-Based Dimensionality Reduction (NLDR)\n		References\n	Tracking of Multiple Objects over Camera Networks with Overlapping and Non-overlapping Views\n		Introduction\n		Related Work\n		Tracking within a Single Camera\n			Detection of Occlusion and Segmentation Errors\n			Measurement Selection via Segmented VOs\n			Measurement Selection via Adaptive Particle Sampling\n			Adaptive Appearance\n		Tracking Across Multiple Cameras\n			Establish Field of View (FOV) Lines\n				Extracting Landmark Points\n				Findling Matching Landmark Points\n				Aligning Two Images\n			Brightness Calibration of Neighboring Cameras\n			Consistent Labeling Across Cameras\n		Experimental Results\n			Performance of Tracking within a Single Camera\n			Performance of Tracking Across Multiple Cameras\n		Conclusion\n		References\n	Toward Robust Online Visual Tracking\n		Introduction\n		Appearance Modeling for Visual Tracking\n			Learning Nonlinear Appearance Manifold\n				Learning Nonlinear Manifold Online\n				Online Update of Submanifold\n			Leveraging Prior Knowledge with Online Learning\n		Learning Detectors Online for Visual Tracking\n			Multiple Instance Learning\n			Learning Detectors with Online Multiple Instance Boosting\n			Articulated Objects\n		Conclusions\n		References\n	Modeling Patterns of Activity and Detecting Abnormal Events with Low-Level Co-occurrences\n		Introduction\n		Context, Overview and Notations\n			Context\n			Overview and Notation\n		Our Method\n			Training Phase\n				Nominal Model\n				Learning the Co-occurrence Matrix\n				A Specific Case for Co-occurrence\n				Complexity Issues & Conditional Independence\n			Observation Phase\n				Abnormal Model\n				Abnormality Detection\n		Experimental Results\n		Conclusion\n		References\n	Use of Context in Video Processing\n		Introduction\n		Case Study: Environment Discovery\n			Environmental Context\n				Camera Priors in Activity Recognition\n				Object Recognition Through User Activities\n			User-Based Context\n				Adapting Domain Knowledge from User Feedback\n		Conclusion\n		References\nSimulation, Graphics, Cognition and Video Networks\n	Virtual Vision\n		Introduction\n		The Case for Virtual Vision\n		Related Work\n		Smart Camera Nodes\n			Synthetic Video\n			Visual Processing\n			Camera Node Behavioral Controller\n		Surveillance Systems\n			Active Camera Scheduling\n			Collaborative Persistent Surveillance\n		Conclusions\n		References\n	Virtualization and Programming Support for Video Sensor Networks with Application to Wireless and Physical Security\n		Motivation\n		Related Work\n			Wireless Intrusion Detection\n			Intrusion Detection Systems\n			Wireless Intrusion Detection Systems\n		SNBench Overview\n		Enabling Wireless Monitoring\n			WifiAlertSensor\n			WifiActivitySensor\n			WifiResponder\n		Deployment Environment\n		Service Programming Primer\n		Wireless Security Services\n		Future Work and Conclusions\n			Wireless Access Lists from Physical Data\n			snBench as a Complete, Turn-Key Network Security Solution\n			In Conclusion\n		References\n	Simulating Human Activities for Synthetic Inputs to Sensor Systems\n		Overview\n		The CAROSA System\n		Related Work\n		Parameterized Representations\n		Resource Management\n		Roles and Groups\n		Scenario Authoring\n		Example Simulation\n		CAROSA Summary\n		Input to Distributed Sensor Networks\n		Summary\n		References\n	Cognitive Sensor Networks\n		Introduction\n		Cognition\n			The Domain Theory Hypothesis\n			Sensor Networks\n			Domain Theory-Based Perception\n		Symmetry Theory in Signal Processing\n		Conclusion\n		References\n	Ubiquitous Displays: A Distributed Network of Active Displays\n		Introduction\n			State of the Art: Centralized Displays\n				Passive Multi-Displays\n				Single Active Displays\n				Bottleneck of Centralized Systems\n			Disruptive Change in Display Metaphor\n				Active Displays\n				Interaction with Environment\n				Interaction with User and Data\n		Initial Progress\n			Distributed Self-Calibration of Planar Display Walls\n			Color Registration Amenable to Parallelization\n			Distributed Interaction with 2D Applications on Planar Display Walls\n			Projector Camera Self Calibration Techniques\n			Registering Constrained Non-Planar Displays Using a Single Uncalibrated Camera\n		Conclusion\n		References\nWireless Video Sensor Networks, Communications and Control\n	Research Challenges for Wireless Multimedia Sensor Networks\n		Introduction\n		Applications of Wireless Multimedia Sensor Networks\n		Network Architecture\n		Factors Influencing the Design of Multimedia Sensor Networks\n		Application Layer\n			Multimedia Encoding Techniques\n				Distributed Video Coding\n				Video Encoding Based on Compressed Sensing\n			Collaborative In-network Processing\n		Transport Layer Protocols\n			UDP Based Protocols\n			TCP and TCP Friendly Schemes for WMSNs\n			Distortion-Minimizing Rate Control\n		Network Layer\n		MAC Layer\n			Contention-Based Protocols\n			Contention-free Single Channel Protocols\n		Physical Layer\n		Conclusions\n		References\n	Camera Control and Geo-Registration for Video Sensor Networks\n		Introduction\n		Related Work\n		PTZ Camera Viewspace Control Model\n		Scene-Based Camera Geo-Registration and Mapping\n		Operational Interface\n		Summary\n		References\n	Persistent Observation of Dynamic Scenes in an Active Camera Network\n		Introduction\n		Technical Rationale\n			Necessity of Collaboration in an Active Camera Network\n			Necessity of a Decentralized Strategy\n			Relation to Previous Work\n		Cooperative Target Acquisition Using Game Theory\n			Motivation for Game-Theoretic Formulation\n			Precise Problem Statement and Notation\n			Game-Theory Fundamentals\n			Choice of Utility Functions\n				Target Utility\n				Global Utility\n				Camera Utility\n			Negotiation Mechanisms\n				Application of SAP Negotiation Mechanism\n		Experimental Results\n		Conclusion\n		References\n	Proactive PTZ Camera Control\n		Introduction\n			Related Work\n		Proactive Camera Control\n			Problem Statement\n			Finding Good State Sequences\n				PTZ Camera Relevance\n				State Sequence Quality\n			Planning\n			Finding an Optimal Sequence\n		Results\n		Conclusions and Future Work\n		References\n	Distributed Consensus Algorithms for Image-Based Localization in Camera Sensor Networks\n		Introduction\n			Chapter Contributions\n			Related Work\n		Review of Average-Consensus Algorithms\n		Distributed Object Localization\n			Consensus on SO(3)\n			Consensus on so(3)\n			Simulation Results\n		Distributed CSN Localization\n			Estimation of the Rotations\n			Estimation of the Translations\n			Complete Estimation\n			Simulation Results\n		Conclusion\n		References\n	Conditional Posterior Cramér-Rao Lower Bound and its Applications in Adaptive Sensor Management\n		Introduction\n			Information Theoretic Measures\n			PCRLB\n		Conditional PCRLB for Recursive Nonlinear Filtering\n			CRLB\n			Unconditional PCRLB\n			Conditional PCRLB\n			Recursive Conditional PCRLB\n		C-PCRLB-Based Sensor Management\n			Adaptive Sensor Selection for Iterative Source Localization\n			Dynamic Sensor Selection for Tracking\n		Applications in Camera Network Management\n		References\nDistributed Embedded Cameras and Real-Time Video Analysis\n	VideoWeb: Optimizing a Wireless Camera Network for Real-time Surveillance\n		Introduction\n		Related Work and Contributions\n		Building the Camera Network\n			Choosing the Type of Network\n			Choosing the Right Camera\n			Choosing and Configuring the Network Hardware\n		The VideoWeb Wireless Camera Network\n		Experiments for Performance Characterization and Optimization of the Video Network\n			Optimizing Camera Configuration\n			Multi-objective Optimization Using Pareto Efficiency\n				Inferiority and Non-Inferiority\n				Data Collection\n			Evaluation Results\n		Conclusions\n		References\n	VideoWeb Dataset for Multi-camera Activities and Non-verbal Communication\n		Introduction\n		Data Collection\n			Purpose and Significance of Data\n			Environment for Data\n			Contents of Data\n				Examples\n			Ground-Truth Annotations\n			Availability of the Data\n		Conclusions\n		References\n	Wide-Area Persistent Airborne Video: Architecture and Challenges\n		Introduction\n		Spatio-temporal Reflectance Variations\n		Wide Aperture Imaging Model of Camera Arrays\n			Seamless Stitchable Camera Arrays\n			Geometric Properties of WFOV Imaging Arrays\n			Physical Considerations Governing Camera-Array-based WFOV Virtual Focal Planes\n		Accommodating Dynamic Variations in Operational Camera Arrays Using Pose Information\n		Summary and Conclusions\n		References\n	Collaborative Face Recognition Using a Network of Embedded Cameras\n		Introduction\n			Contributions and Main Results\n			Outline of the Paper\n		Related Work\n		Experimental Setup\n			System Model\n			Assembly of Camera Platform\n			Assembly of Embedded Camera Network\n			Software Implementation\n			Experimentation\n		System Performance\n			Network Performance\n			Face Recognition Performance\n			Real-time Capability\n		Conclusions and Future Work\n		References\n	SATware: A Semantic Approach for Building Sentient Spaces\n		Introduction\n		SATware: An Middleware Framework for Sentient Spaces\n		A Programming Model for Pervasive Applications\n			Virtual Sensors: Bridging Application Needs to Raw Sensor Streams\n			Query Processing in SATware\n		Supporting Scalability through Semantic Scheduling\n		Supporting Robustness through Sensor Recalibration\n		Conclusions\n		References\nApplications of Distributed Video Networks\n	Video Analytics for Force Protection\n		Aerial Video Analysis\n			Stabilization\n			Object Detection\n			Tracking\n			Super-resolution\n		Tracking from Fixed Ground Based Cameras\n		Person Detection from Moving Platforms\n		Biometrics at a Distance\n			Face Capture and Recognition\n			Iris Recognition\n			Whole-Body Re-Identification\n		Facial Analysis\n		Summary\n		References\n	Recognizing Activity Structures in Massive Numbers of Simple Events Over Large Areas\n		Introduction\n		Spatial Structure\n		Temporal Structure\n		Event-Linkage Structure\n		Short Event-Sequence Structure\n		Network Structure\n		Summary\n		References\n	Distributed Sensor Networks for Visual Surveillance\n		Introduction\n		Technical Challenges in Large Sensor Networks\n		System Design and Components\n			Auto Camera Calibration and Geo-Registration\n			Video Processing\n			Efficient Processing of High-Resolution Imagery\n			Context Learning\n			Environmental Conditions\n			Scene Types and Elements\n			Target Property Models\n			Data Fusion and Event Inference\n			User Interface\n		Results\n		References\n	Ascertaining Human Identity in Night Environments\n		Introduction\n		Color-NIR Cross-Spectral Iris Matching\n			Multispectral Iris Dataset and Data Used in Simulations\n			Proposed Predictive Model\n			Recognition Performance\n		Short Wave Infrared Face Verification\n			SWIR Data Collection\n			Methodology\n				Preprocessing and Normalization\n				Matching Experiments\n			Results\n				SWIR Face Verification\n		Gait Curves for Human Recognition in a Night-Time Environment\n			Methodology\n				Preprocessing and Silhouette Extraction\n				Spatio-temporal Feature Extraction\n			Experiments and Results\n				Classification of Human Gait\n			Discussion\n		Soft Biometrics-Body Measurement\n		Summary\n		References\nEducational Opportunities and Curriculum Development\n	Educational Opportunities in Video Sensor Networks\n		Introduction\n		Computational Sensor Networks\n		Engineering Background for Video Sensor Networks\n		Course Organization\n		Support Technology for Instruction\n		Conclusion\n		Recommended Courses and Topics\n			Machine Vision\n			Sensor Networks\n			Hardware Systems\n			Software Systems\n		References\nIndex