2003 Volume 69 Issue 680 Pages 1011-1018
This paper proposes a seamless tracking system that is a real time visual surveillance system for detecting and tracking people and monitoring their activities in a wide area environment with multiple overhead cameras. This system is composed of multiple vision systems with overhead cameras and a database server. AIl vision systems are connected with a database server through a computer network in order to upload and download some information of a detected human such as hair color, cloth color, and human location. A vision system can identifies people previously detected by a vision system, checking corresponding features and position stored in the database. AIl people are therefore tracked in a wide area environment. We propose the seamless tracking system and the ways to detect and to identify those people and then show their performance using our developed system.