Tracking and modeling people in video sequences

Tracking and modeling people from video sequences has become an increasingl y important research topic, with applications including animation, surveill ance, and sports medicine. In this paper, we propose a model-based 3-D appr oach to recovering both body shape and motion. It takes advantage of a soph isticated animation model to achieve both robustness and realism. Stereo se quences of people in motion serve as input to our system. From these, we ex tract a 2 1/2-D description of the scene and, optionally, silhouette edges. We propose an integrated framework to ft the model and to track the person 's motion. The environment does not have to be engineered. We recover not o nly the motion but also a full animation model closely resembling the subje ct. We present results of our system on real sequences and we show the gene ric model adjusting to the person and following various kinds of motion. (C ) 2001 Academic Press.