Double spline muscle models for human face animation based on video analysis

Facial animation has been widely used in different areas such as entertainment and teleconferencing. Using video data to control facial animation is an interesting and difficult problem. In this paper, a novel double B-spline (NURBS) muscle system is proposed to simulate a 3D facial expression and talking animation based on video images. This system can simulate many different expressions of the elastic human face. Each muscle is constructed by two non-uniform rational B-spline curves, which better approximate the real muscle system. There are 14 feature points to represent facial expressions and we apply a lip contour extraction technique to determine lip shapes.