Perception and emotion modeling of avatars in IVE

In intelligent virtual environment (IVE), it is always a challenge research issue for the virtual actor (avatar) to have the ability of visual perception and response to the virtual world. And facial expression and walk behavior of avatars play an important role in IVE. In our method, the avatar is considered as an agent, it recognizes 3D objects in the environment and calculates the direction of the moving objects by using the avatar's perception sensor, avatar's facial expression is rule-based transition. Avatar walk behavior is simulated by skeleton model and is denoted by seven-motion sequence. Facial expression of individuals can be simulated after expression cloning and face 3D model reconstruction based on facial muscular model. Experimental results show the efficiency of our approach.