Active-vision system for multi-target surveillance

This paper presents a novel agent-based method for the dynamic, coordinated selection and positioning of groups of active-vision cameras for the surveillance of multiple object-of-interests as they travel through a cluttered environment with a priori unknown trajectories. The proposed dispatching algorithm has been tested via simulations and implemented on an experimental prototype setup for automated facial recognition. Both analyses have shown that the use of dynamic sensors along with an on-line dispatching algorithm can tangibly improve the performance of a surveillance system.