An Automatic MPEG-4 Based Realistic 3D Facial Animation Method

Along with the quick development of computer technology, especially the multimedia technology, the research of humanoid human-computer interaction interface comes to the focus. Three-dimensional computer facial animation is an important way for human-computer interaction, and it can be applied to various areas of virtual reality environments such as virtual presenter, net meeting, videophone, remote education, remote medical diagnosis, movie effects and games etc. Since the MPEG-4 based facial animation system has the advantage of little data, computationally cheap and can perform real time animation, it receive more and more attention in recent years. The most important advantage of the MPEG-4 based facial animation system is that it's a much more universal facial animation system. This dissertation focuses mainly on the capture of real human facial animation data and the automatic MPEG-4 based arbitrary topology facial animation method.The study areas in 3D facial animation include facial modeling and facial animation. The facial animation research includes study of facial motion capture and facial animation control strategy. This paper mainly deals with the problem of facial animation data capture and facial control strategy in the MPEG-4 based facial animation system.Facial motion capture plays an important role in any facial animation system. In this paper we implemented the method to capture real facial motion data by making marks on facial images, and then transform these motion data into the form of the MPEG-4 FAPs (Facial animation parameter).The key point in a MPEG-4 based facial animation system is the construction of the FaceDefTables corresponding to the specific facial model. In this paper, by using the ASM (Active shape model) and texture mapping, we proposed an automatic method to locate feature points on 3D facial models. Then based on the 3D feature points on the 3D facial model, we construct FaceDefTables for arbitrary topology facial model automatically.We also make our facial animation system more automatic than previous by automatic add of the lip line and the automatic fix of the wrong motion factor at the mouth region.