With the continuous extension of virtual human technology's Application field, the virtual human technology is also updated. as an important part of human virtual technology, the motion control of virtual human is improving too. However, every virtual human technology has its own advantages and disadvantages.In this paper, after introducing and analyzing the main types of the existing virtual human models and the ways of virtual human motion control, we present a hierarchical, scalable and controllable three-dimension virtual human model from the point of view that raising reusability and fidelity of virtual human technology. On the basis of that model, we also present a new approach based on the standard related to the resolution of human being eyes to controlling a three-dimension virtual human's motion. This method tells us how to control a virtual human's motion according to high-performance ability of human being's eye. It employs the technology called "skinned skeletal animation" as the mean of the output resulted from our control of virtual human. It also takes both the advantages of motion capture technology and parametric key frame technology. We tested and verified our new model and method using a human-like role model. Our method does better than motion capture technology from efficiency and precision to effect. All these tell us our method is an effective, realistic way for virtual human motion control. |