From Regular Images to Animated Heads: A Least Squares Approach
P. Fun · 2006
1 I n t r o d u c t i o n In this paper, we show tha t we can effectively fit a complex head an imat ion model , including ears and hair, to image da ta obtained using simple video or CCD cameras as opposed to sophisticated sensors such as laser range finders. In recent years much work has been devoted to the modell ing of faces f rom image and range data. There are m a n y effective approaches to recovering face geometry. They rely on stereo [5], shading [14], s t ructured light [19], silhouettes [20] or low-intensity lasers. Some of these systems such as the C3D t'~l or the Cyberware tm scanner are commercial ly available. However, recovering a head as a simple t r iangulated mesh does not suffice: To an imate the face, one mus t fur ther fit an actual an imat ion model to the data. A u t o m a t e d approaches to this task can be roughly classified into the following