A Video-Rate Stereo System by Integration of Two Algorithms with/without Occlusion Handling

This paper proposes a novel architecture for video-rate stereo machine. It is unfeasible to directly implement a video-rate machine with occlusion handling mechanism, because the computational cost for the correspondence search is higher than that of a simple polynocular stereo algorithm without occlusion handling. Our idea is to integrate two algorithms withlwithout occlusion handling mechanism in a framework of video-rate stereo machine. By switching two depth maps using motion information, the video-rate depth map is output for dynamic regions, while for static regions is given the dense depth map with sharp object boundaries. We have implemented a prototype system by using a commercially available video-rate stereo machine and a PC with an occlusion detectable algorithm on it.