Abstract
Pose estimation in the context of human motion analysis is the process of approximating the body configuration in each frame of a motion sequence. We propose a novel pose estimation method based on fitting a skeletal model to tree structures built from skeletonised visual hulls reconstructed from multi-view video. The pose is estimated independently in each frame, hence the method can recover from errors in previous frames, which overcomes some problems of tracking. Publically available datasets were used to evaluate the method. On real data the method performs at a framerate of similar to 14 fps. Using synthetic data the positions of the joints were determined with a mean error of similar to 6 cm.