Assigning cameras to subjects in video surveillance systems

Abstract

Abstract-We consider the problem of tracking multiple agents moving amongst obstacles, using multiple cameras. Given an environment with obstacles, and many people moving through it, we construct a separate narrow field of view video for as many people as possible, by stitching together video segments from multiple cameras over time. We employ a novel approach to assign cameras to people as a function of time, with camera switches when needed. The problem is modeled as a bipartite graph and the solution corresponds to a maximum matching. As people move, the solution is efficiently updated by computing an augmenting path rather than by solving for a new matching. This reduces computation time by an order of magnitude. In addition, solving for the shortest augmenting path minimizes the number of camera switches at each update. When not all people can be covered by the available cameras, we cluster as many people as possible into small groups, then assign cameras to groups using a minimum cost matching algorithm. We test our method using numerous runs from different simulators.

Viewpoint Selection for Human Actions

“…Most of the cameras in the cluster do not overlap and hence the main goal is tracking the person as he moves from one camera view to the other. In Krahnstoever et al (2008), El-Alfy et al (2009 methods were proposed for selecting the camera that provides the best view for recognizing the identity of a viewed person. This requires mostly face visibility.…”

Exact algorithms for non-overlapping 2-frame problem with non-partial coverage for networked robotic cameras

“…While the wide-angle static camera monitors the scene and detects pre-defined individual human activities (e.g., loitering), the PTZ camera takes high-resolution images of the human for close-up observation. El-Alfy et al [6] model the subject-camera assignment issue for a PTZ camera network as a maximum matching problem in a bipartite graph and group the subjects using heuristics when there are less camera than subjects. Different from these existing work, the solution to the p-frame problem can be applied to optimally control PTZ camera parameters such that the camera coverage-resolution tradeoff is achieved by maximizing the satisfaction level of the observation to all objects.…”