Abstract:
Hand pose modeling is one of the most challenging research areas in computer animation because of its complicated motion. Since hand consists of not less than 20 degrees of freedom which is hard to model, the ways to solve this problem require to construct a large multi-dimensional database of poses which obtained by using special equipment. Therefore, both time and cost are much consumed. In this research, we propose a method of 3D hand pose modeling based on new set of constraints and optimization framework to reduce the ambiguity between 2D joint location and 3D hand poses. We propose a new constraint that is symmetry constraint and rigid body constraint for hand pose modeling. Our method does not need prior knowledge of the bone length and camera parameters. Furthermore, a large high-dimension database used for searching is not needed to be constructed, which is considerably reduced the cost. From this research the error of bone length estimation obtained from our experiments are around 31%, however, the system can be archived.