Surveillance Camera Coordination Through Distributed Scheduling

A challenge to scaling a video surveillance system is the amount of human supervision required for control of the cameras. In this paper we consider the problem of coordinating a network of video cameras for the purpose of identifying people. We pose the problem as a machine scheduling problem where each person is a job that should be scheduled before a deadline. To ensure scalability, we propose a distributed algorithm that only depends on neighbor to neighbor communication. We compare the performance of this algorithm to a localized scheduling approach.