Automatic 3D CBIR on kinematical human motion
Chun-Hong Huang, Ching-Sheng Wang, Mengliang Yu · 2007
Abstract: Automatic retrieval of human body motion is one of the most challenging issues of video and virtual reality research. In this paper, we propose a human motion retrieval system, which allows users to retrieve 3D kinematical motions. The system includes two major components for motion analysis and comparison. The first is a recognition unit of motion types which is based on Skeleton Discrimination Tree. The second is a unit for synchronization and motion similarity comparison. The comparative approach is based on mutative dynamic programming that considers the degree of the included angles of the vectors belonging to individual feature tracks. With Motion Capture Camera and animations represented by VRML, it is possible to automatically summarize, analyze and adjust the 3D motions of a real person. Users may provide a 3D VRML human motion object and find the similar human motions via our system.