Interactive facial animation system

Abstract

A generic and interactive facial animation system for synthesizing the facial expres- sions of human and animals is developed. Effective paradigms based on parameterised approaches are developed and integrated into the facial animation system. An effective and practical tools are developed for facilitating the workflow of generating facial anima- tion based on blend shape and set driven key approaches. We proposed a novel muscle deformation method, namely neuroEnveloping. The main idea relies on employing artificial neural networks to learn how to synthesize deformable muscle from example shapes of facial expressions paired with skeleton and influence objects.