Bayesian Multi-Camera Surveillance

The task of multi-camera surveillance is to reconstruct the paths taken by all moving objects that are temporarily visible from multiple non-overlapping cameras. We present a Bayesian formalization of this task, where the optimal solution is the set of object paths with the highest posterior probability given the observed data. We show how to efficiently approximate the maximum a posteriori solution by linear programming, and present initial experimental results. 1 Multi-camera surveillance Video surveillance in a large or complex environment requires the use of multiple cameras. In this paper we address a particular task that we call multicamera surveillance (MCS). The multi-camera surveillance task arises in an environment with moving objects that is monitored by multiple non-overlapping cameras, such as an office building with pedestrians, or a set of highways. The task is to reconstruct the paths taken by all objects that were visible during the observation period, despite the fac...

• Type: Conference
• Year: 1999
• Where: CVPR
• Authors: Vera Kettnaker, Ramin Zabih
• Keywords: Computer Vision, CVPR 1999, MCS Task, Multi-camera Surveillance Task, Multiple Non-overlapping Cameras, Object Paths, Surveillance Video Surveillance