Emotion model of interactive virtual humans on the basis of MDP

Abstract

Emotion plays an essential role in the adaptation and social communication of organisms. Similarly, an appropriately timed and clearly expressed emotion is a central requirement for believable interactive virtual humans. Presently, incorporating emotion into virtual humans has gained increasing attention in the academia and industry. This strong interest is driven by a wide spectrum of promising applications in many areas such as virtual reality, e-learning, entertainment, etc. This paper introduces an emotion model of artificial psychology, in which the transition of emotion can be viewed as a Markov process and the relation of emotion, external incentive and personality can be described by a Markov decision process (MDP). In order to demonstrate the approach, this paper integrates the emotion model into a system composed of voice recognition and a realistic facial model. Thus, the model could be used for generating a variety of emotional expressions of autonomous, interactive virtual human beings.