Seamless tracking system with multiple cameras

This paper proposes a "seamless tracking system", which is a real-time visual surveillance system for detecting and tracking people and monitoring their activities in a wide-area environment. This system is constructed from many vision systems, and they are connected through a computer network. Each vision system consists of a camera and an image processor. They recognize target positions in the wide-area environment and exchange observed information with each other in order to track a specific person. In this paper, we propose a method for human tracking and coordination of the entire vision system.