An occlusion metric for selecting robust camera configurations

Vision based tracking systems for surveillance and motion capture rely on a set of cameras to sense the environment. The exact placement or configuration of these cameras can have a profound affect on the quality of tracking which is achievable. Although several factors contribute, occlusion due to moving objects within the scene itself is often the dominant source of tracking error. This work introduces a configuration quality metric based on the likelihood of dynamic occlusion. Since the exact geometry of occluders can not be known a priori, we use a probabilistic model of occlusion. This model is extensively evaluated experimentally using hundreds of different camera configurations and found to correlate very closely with the actual probability of feature occlusion.

• Type: Journal
• Year: 2008
• Where: MVA
• Authors: Xing Chen, James Davis
• Keywords: Camera, Exact Placement, Motion Capture, MVA 2007, MVA 2008