Surveillance camera scheduling: A virtual vision approach,” VSSN ’05

We present a surveillance system, comprising wide field-of-view (FOV) passive cameras and pan/tilt/zoom (PTZ) active cameras, which automatically captures and labels high-resolution videos of pedestrians as they move through a designated area. A wide-FOV stationary camera can track multiple pedestrians, while any PTZ active camera can capture high-quality videos of a single pedestrian at a time. We propose a multi-camera control strategy that com-bines information gathered by the wide-FOV cameras with weighted round-robin scheduling to guide the available PTZ cameras, such that each pedestrian is viewed by at least one active camera during their stay in the designated area. A distinctive centerpiece of our work is the exploitation of a vi-sually and behaviorally realistic virtual environment simulator for the development and testing of surveillance systems. Our research would be more or less infeasible in the real world given the im-pediments to deploying and experimenting with an appropriately complex camera sensor network in a large public space the size of, say, a train station. In particular, we demonstrate our surveil-lance system in a virtual train station environment populated by autonomous, lifelike virtual pedestrians, wherein easily reconfig-urable virtual cameras generate synthetic video feeds that emu-late those generated by real surveillance cameras monitoring richly populated public spaces