Reproduction of a Human Motion based on Marker-less Motion Capture

This paper presents a real-time marker-less motion capture system, whose purpose is to do seamless mapping objects in the real world into virtual environments easily. In general, virtual environment applications such as man-machine seamless interaction require the system to estimate motion parameters at real-time for natural objects such as human bodies. To achieve this requirement, we have developed a marker-less motion capture system which realizes multiple camera fusion and reconstructs human whole body motion parameters by real-time inverse kinematics. Advantage of such vision-based system is that it is possible to acquire the other scene parameters such as shape and surface properties at the same time, using the same equipments in measuring motion. In this paper, we have implemented a marker-less motion capture and reconstruction system and have demonstrated the system works at real-time and online on PC-cluster