【摘要】 在人脸重构系统与智能虚拟环境系统的研究基础上,提出了一种基于Agent的建立虚拟人感知能力与行为选择模型的新方法。该方法把虚拟人(Avatar)当成一个Agent,它在虚拟场景中能够识别三维物体以及利用传感器计算物体的运动方向。Avatar的行走采用骨架模型来进行仿真,并且其行走行为由七个基本动作行为组成。实验结果显示新方法是具有较好的效果。更多还原

【Abstract】 Based on the research of 3D facial model reconstruction and intelligent virtual entrainment, a new perception and autonomous behavior modeling approach for Avatar was proposed. In the new method, the Avatar was considered as an Agent,which could recognize 3D objects in the environment and calculate the direction of the moving objects by using the Avatar’s perception sensor. Avatar walk behavior was simulated by skeleton model and denoted by seven motion sequences. Experiment results show that the new approach is more efficiency.更多还原