Skip to Main Content
One of the fundamental requirements for visual surveillance with smart camera networks is the correct association of camera's observations with the track of objects under tracking. Most of the current systems are centralized in that the observations on all cameras need to be transmitted to a central server where some data association algorithm is running. In this paper, we propose a distributed data association approach in which the data association inference is achieved by propagation of beliefs of the association variables between neighboring cameras. We also incorporate the distributed inference into distributed EM framework to solve the problem of data association and object appearance learning simultaneously in a complete distributed manner. The proposed method is tested on artificial data and on real world observations collected by a camera networks in an office building.