| This thesis is an investigation of 3-D monocular visual tracking of human limbs in unconstrained environments. The 3-D visual tracking of human limbs is fundamental to a wide array of computer vision applications including gesture recognition, interactive entertainment, biomechanical analysis, vehicle driver monitoring, and electronic surveillance. The problem of limb tracking is complicated by issues of occlusion, depth ambiguities, rotational ambiguities, and high levels of noise caused by loose fitting clothing. Many conventional visual limb tracking systems have attempted to solve the problem by introducing multiple cameras (i.e. stereo vision) to reduce occlusion and depth ambiguities. As well, most tracking systems have simplified the problem by using constrained environments, such as an assumption of a static background, forcing the user to wear tight fitting clothing against a contrasting background, or relaxing the real-time operating conditions. In this thesis we attempt to solve the 3-D limb tracking problem using only monocular imagery (a single 2-D video source) in largely unconstrained environments. (Abstract shortened by UMI.)... |
