Automatic construction of 3D animatable facial avatars

Rigging for facial animation is an important but time-consuming task, which generally requires experienced artists with knowledge of facial anatomy. In this paper, we investigate whether it is possible to produce a good animatable avatar automatically, given only a 3D static triangle mesh of the head. An automatic mechanism is devised for constructing multi-layer animatable facial avatars for unseen faces. We evaluate our technique with a variety of models, and give a quantitative analysis of the constructed results. We also designed and conducted a user study for evaluating the perceived quality of the generated expressive animations. The results demonstrate that our method is an appropriate tool for naïve users to customize their personal 3D avatars. Copyright © 2010 John Wiley & Sons, Ltd. In this paper, we investigate whether it is possible to produce a good animatable model of a head automatically, given only a 3D static triangle mesh of the head. An automatic mechanism is devised for constructing multi-layer animatable facial avatars for unseen faces. We demonstrate that this method is suitable for naive users who wish to customise 3D avatars for applications such as on-line games and chat rooms.