Techniques for modelling and training multimedia expressive talking heads

This paper describes methods for modelling and training emotional talking heads of virtual actors suitable for computer animated movies and other applications. Models, which can be used to generate and to combine emotional expressions, visual speech, and facial gestures according to a script specifying high-level attributes, are initially presented. The advantage of such techniques is the ability to automatically generate emotional visual speech of synthetic talking faces in a more user-friendly manner with the aid of high level scripts and parameters. The latter sections of the paper present an overview of a virtual actor system, designed to automate some of these animation tasks. An expert system using a fuzzy knowledge-based control system is used to realize the automated system. Fuzzy linguistic rules are used to give virtual actors knowledge of the appropriate emotions and gestures to use in different situations of a synthetic movie. Theories of emotion, personality, dialogue, and acting, as well as empirical evidence are incorporated into the knowledge bases to produce convincing results.