From the streets of London to subway stations in New York City, hundreds of thousands of surveillance cameras ubiquitously collect hundreds of thousands of videos, often running 24/7. How can such vast volumes of video data be stored, analyzed, indexed, and searched? How can advanced video analysis and systems autonomously recognize people and detect targeted activities real-time? Collating and presenting the latest information Intelligent Video Surveillance: Systems and Technology explores these issues, from fundamentals principle to algorithmic design and system implementation.
An Integrated discussion of key research and applications
Written and edited by a collection of industry experts, the book presents state-of-the-art technologies and systems in intelligent video surveillance. The book integrates key research, design, and implementation themes of intelligent video surveillance systems and technology into one comprehensive reference. The chapters cover the computational principles behind the technologies and systems and include system implementation issues as well as examples of successful applications of these technologies.
Builds a foundation for future developments
Changing appearance caused by changing viewpoints, illumination, expression, and movement, self/cross body occlusion, modeling of cluttered background capable of efficient background subtraction for object detection, and spatial and temporal alignment of multiple cameras are just a few of the challenges that remain in further developing and refining intelligent video surveillance technology and systems. Fully illustrated with line art, tables, and photographs demonstrating the collected video and results obtained using the related algorithms, including a color plate section, the book provides a high-level blueprint for advances and insights into future directions of the field.
Table of Contents
Background, Introdution and Foundation
Fundamentals for Video Surveillance
Case Study: IBM Smart Surveillance System, R. Feris, A. Hampapur, Y. Zhai, R. Bobbitt, L. Brown, D. Vaquero, Y.-l. Tian, H. Liu, and M.-T. Sun
Detection and Tracking
Adaptive Background Modeling and Subtraction: A Density-based Approach with Multiple Features, B. Han and L.S. Davis
Pedestrian Detection and Tracking, B.Wu and R. Nevatia
Vehicle Tracking and Recognition, G. Fan, X. Fan, V. Venkataraman and J.P. Havlicek
Articulated Human Motion Tracking in Low-Dimensional Latent Spaces, G. Qian and F. Guo
Human Action Recognition, X. Xu and B. Li
Complex Activity Recognition, J. Muncaster and Y. Ma
Multi-level Human Interaction Recognition, S. Park and M.M. Trivedi
Multi-Camera Calibration and Trajectory Fusion, N. Anjum and A. Cavallaro
Dynamic Camera Assignment and Handoff, B. Bhanu and Y. Li
Quality Based Multi-Sensor Fusion, L. Snidaro, I. Visentini, and G.L. Foresti
Systems and Applications
Attribute-Based People Search, D.A. Vaquero, R.S. Feris, L. Brown, A. Hampapur, and M. Turk
Soft Biometrics for Video Surveillance, Y. Fu, G. Guo, and T.S. Huang
Moving Platform System in Geo-coordinates, Q. Yu, Y. Lin and G. Medioni
Urban 3D Reconstruction, J.-M. Frahm and M. Pollefeys
Yunqian Ma is a Principal Scientist for Honeywell Labs. Gang Qian is a Professor of Electrical Engineering at Arizona State University.