This paper describes a new method for tracking of a human body in 3D motion by using constraints imposed on the body from the scene. An image-based approach for tracking exclusively uses a geometrical model of the human body. Since the model usually has a large number of degrees of freedom (DOF), a chance to be corrupted by noise increases during the tracking process, and the tracking may fall in an ill-posed problem. To cope with this problem, we pay our attention to that a human body cannot move freely, and usually receive some constraints a from the scene. The new method uses constraints imposed on position, velocity and acceleration of the part of the body from the scene. These constraints can reduce the DOF of the model. This reduction guarantees the tracking problem to be a well-posed problem, and prevents tracking errors by noise. Experiments with real image sequences support a precise tracking of the body.

The model-based approaches for tracking of human bodies in image sequences can be categorised into two types; fitting model to body frame by frame, and accumulating estimated pose displacements in successive frames after model fitting at the initial frame. The latter has an inherent drawback as accumulation of tracking errors while the one has a great advantage as small computational efforts compared with the former. This paper proposes a new method which can correct the tracking errors by propagation from fitting model to body at a few key-frames. The propagation makes it possible to establish tracking of bodies under occlusion. Capturing the actor's motions in real old movies is presented.