Background | Virtual Vision | Movies (Smart cameras) | Acknowledgements | References
Terzopoulos [1] proposed a Virtual Vision approach to designing surveillance systems using a virtual train station environment populated by fully autonomous, life like pedestrians that perform various activities [2]. Within this environment, virtual cameras generate synthetic video feeds. The video streams emulate those generated by real surveillance cameras, and low-level image processing mimics the performance characteristics of a state-of-the-art surveillance video system.
It is an on going project and the related publications are available here.
|
|
The movies are either in mp4 or divx format. A free video player for Windows, OSX, and numerous other operating systems is available here.
The following movies show our on going work in virtual vision. Here, we show virtual camera sensor networks comprising passive wide field-of-view and active pan/tilt/zoom cameras performing strategic visual surveillance tasks. The cameras are deployed in the virtual Penn station.
The research reported herein was made possible in part by a grant from the Defence Advanced Research Projects Agency (DARPA) of the Department of Defence. We thank Dr. Tom Strat, formerly of DARPA, for his generous support and encouragement. We also thank Wei Shao and Mauricio Plaza-Villegas for their invaluable contributions to the implementation of the Penn Station simulator.
[1] D. Terzopoulos, “Perceptive agents and systems in virtual reality,” in Proc. 10th ACM Symposium on Virtual Reality Software and Technology, (Osaka, Japan), pp.1–3, Oct. 2003.
[2] W. Shao and D.Terzopoulos, “Autonomous pedestrians,” in Proc. ACMSIGGRAPH/Eurographics Symposium on Computer Animation, (LosAngeles, CA), pp. 19–28, July 2005.