Synthesis of Combined Facial Expressions Using Anatomy-based Model

It is still a difficult problem for the synthesis of combined facial expressions in realistic facial animation. In this paper, we present an approach to synthesize combined facial expressions. The approach exploits a uniform model which is based on the anatomical structure of human face to record the facial expressions as facial vectors. Then different facial expressions are synthesized by combining different facial vectors. In this paper, we also develop an action interpretation system for the facial animation. This system defines some primitives of facial states and some rules for the combination of primitives. So the continuous and complex facial animations can be made by combining the primitives and the interval time