Emotional modeling for a vision-based virtual character

Emotion intelligence is an important component to enhance the credibility of virtual character's behavior. A virtual character who can perceive and timely to express the appropriate emotions can make the communication between a visual character and a human user more effective and natural. Based on emotional psychology and artificial psychology, this paper simulates the interaction between human emotions, needs and personalities, proposes a new generation method of virtual character's autonomous emotion, and constructs a vision-Based interactive system composed of a facial expression recognition module and a realistic human face model. The results show that the proposed method can make the virtual character produce a more natural emotional reaction in the interaction, and provide an effective method for designing the virtual character's autonomous emotions in the interactive environment.