An overview of a probabilistic tracker for multiple cooperative tracking agents

Abstract

An overview of a probabilistic cooperative tracking approach is presented in this paper. First, a new tutorial-like detailed explanation of the condensation algorithm by Isard and Blake, (1998) is described. Then we apply the probabilistic tracker to track an object (easily extendable to multiple objects) according to multiple degrees of freedom of the cameras that are able to pan, tilt and zoom. To increase the robustness of the tracking system we extend the one camera tracking method to multiple camera case and each camera is considered as an agent that can communicate with a central unit or it can act based on its own decision. Each camera will gain a level of reliability during the tracking that is used in probabilistic tracking method to improve the performance