In this project, the authors facilitated achieving a common goal in many forensics vision applications, which is to identify and track human objects with distinctive visual features or "tags".
In this paper, the authors made two contributions to this "visual tagging" problem. First, they propose a general framework for camera placement. This framework can measure the performance of any particular camera placement using simulation method. The optimal placement strategy can be obtained by iterative grid-based linear programming. Second, the authors focused on tracking specific, colored tags used in a privacy-protecting visual surveillance network. By building a color classifier for tag detection and using epipolar geometry between multiple cameras for occlusion handling, the proposed system can identify, track, and visually obfuscate individuals whose privacy in the surveillance video needs to be protected. (Publisher abstract provided)
Downloads
Similar Publications
- The Fluid Dynamics of Forensic Bloodstain Analysis: Droplet Impact on Inclined Surfaces
- Development of a spectral X-ray fluorescence database to strengthen the scientific foundations for the forensic analysis and interpretation of modern soda-lime glass
- Development of a matrix-matched standard for the elemental analysis of human hair by LA-ICP-MS