A Multi-Camera Visual Surveillance System for Tracking of Reoccurrences of People

This paper describes a software system to track the reoccurrences of objects in multi-camera visual surveillance. Specifically, it is designed for after-event tracking of people to aid in a typical investigation of events occurring in a certain location at a certain time. This is a nontrivial problem because of several aspects that influence the appearance of scenes and people, such as changes in viewpoints, lighting conditions, shadow, occlusion, and weather conditions. Another challenge, which is the focus of this paper, is to integrate different required components into a complete working system, namely (i) motion detection, (ii) object classification, (iii) object modeling and matching, and (iv) interactive retrieval and visualization. We have designed and implemented a robust system consisting of state-of-the-art technologies in each component. We performed experiments with the system on a real-life dataset gathered from 12 street surveillance cameras over two hours in a city area. The experiments showed promising results in retrieving the reoccurrences of four target subjects.

[1]  Arnold W. M. Smeulders,et al.  Object recognition with uncertain geometry and uncertain part detection , 2005, Comput. Vis. Image Underst..

[2]  T. List,et al.  Comparison of target detection algorithms using adaptive background models , 2005, 2005 IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance.

[3]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[4]  Larry S. Davis,et al.  Efficient Kernel Density Estimation Using the Fast Gauss Transform with Applications to Color Modeling and Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[6]  Cordelia Schmid,et al.  Evaluation of Interest Point Detectors , 2000, International Journal of Computer Vision.

[7]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[8]  W. Eric L. Grimson,et al.  Learning Patterns of Activity Using Real-Time Tracking , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[10]  Larry S. Davis,et al.  Efficient non-parametric adaptive color modeling using fast Gauss transform , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[11]  Larry S. Davis,et al.  W4: Real-Time Surveillance of People and Their Activities , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  A. Leonardis,et al.  On-line Conservative Learning for Person Detection , 2005, 2005 IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance.

[13]  Paul A. Viola,et al.  Robust Real-Time Face Detection , 2001, International Journal of Computer Vision.

[14]  Andrew Zisserman,et al.  Video Google: a text retrieval approach to object matching in videos , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[15]  Arnold W. M. Smeulders,et al.  Color-based object recognition , 1997, Pattern Recognit..