|
Abstract : |
Recently, there has been a movement in computer vision from the processing of static images to the processing of video sequences. Current research has begun to investigate the recognition of human activities taking place in the scene. Applications such as video database, virtual reality interfaces, smart surveillance systems all have in common to track and interpret human activities. We propose a low-cost PC based real-time visual surveillance system, called W 4, for tracking people and their body parts, and monitoring their activities in monochromatic and stereo imagery. It operates on grayscale video imagery, or on video imagery from an infrared camera. Unlike many systems for tracking people, our system make no use of color cue. Instead W 4 employs a combination of shape analysis, robust tracking techniques, silhouette based body model to locate and track the people and understand the interaction between people and objects- e.g., people exchanging objects, leaving objects in the scene. A subsequent system, W 4 S,integrated real-time stereo computation into W 4. Incorporation of stereo has allowed us to overcome the di culties that W 4 encountered with sudden illumination changes, shadow and occlusion which makes tracking much harder in intensity images. A new silhouette-based body model Ghost is described to determine the location of body parts while the people are in generic postures. It is a combination of a hierarchical body pose estimation, a convex hull analysis of the silhouette, and a partial mapping from the body parts to the silhouette segments using a distance transform method that incorporates the topology of the human body. Future research is outlined at the end of proposal. 1, |
