Facial animation reflecting personal characteristics by automatic head modeling and facial muscle adjustment

Abstract

We propose a new automatic character modeling system which can generate an individualized head model only from a facial range scan data and an individualized facial animation with expression change. The head modeling system consists of two core modules: the head modeling module which can generate a head model from a personal facial range scan data using automatic mesh completion, and the key shape generation module which can generate key shapes for the generated head model based on physics-based facial muscle simulation with a personal muscle layout estimated from subject's facial expression videos. As a result, we can generate a head model which can synthesize facial expressions and impression similar to the target person. The experimental result shows that we archive to synthesize CG characters that subjects can identify themselves with 84%. Therefore, we conclude that our head modeling system is effective to games and entertainment systems like the Future Cast.