High-level Human Figure Action Control for Vision-based Real-time Interaction

This paper presents an real-time interaction system which consists of human motion tracking using skin-color-based Gaussian blobs and human motion synthesis based on real-time inverse kinematics. Our purpose is to do seam-less mapping of human action in the real world into virtual environments. In general, virtual environment applications such as smart interaction require real-time human motion tracking without special devices or markers. For the sake of realization of smart interaction, we assume that virtual objects existing in virtual environments can afford human figure action, that is, the virtual environments can provide action information for human figure model, or avatar. In this paper, we demonstrate a real-time, on-line and smart desktop interaction which realizes high-level action recog-nition through interaction with virtual objects. 1