Research On The Technique Of Face Modeling Based On Image

Building realistic 3D face models is a challenging task in the field of computer graphics. With the development of virtual reality, movie making, visual telephone, etc, people are paying more attention to the face modeling area. Aiming at the task of face modeling, this thesis focuses on the technique of face modeling based on image. The research work can be organized in the following aspects:1. The thesis elaborates and compares all kinds of techniques in the field of face modeling and animation, and then introduces five representative face modeling and animation systems.2. Differed from the conventional method that uses the other 3D modeling software, the thesis adopts the surface restoring method based on CT data to get the generic face model. Our method can get better data and makes it easy to deal with.3. The thesis studies the face modeling technique based on frontal and side facial images. On the basis of facial structure, we define facial feature points that show not only the whole outline of a face, but also the local character of the facial components. According to the data structure of the generic face model, the thesis brings forward a simple method of 3D point marking to make the marking results accurate. The thesis also studies the method of deriving the 3D coordinate of the feature point from the 2D images. For the modification from the generic face model to the given face model, the thesis adopts an improved method of interpolation based on radial basis functions and gets a good result.4. Combining with the procedure of facial modeling, the thesis improves the texture mapping method of WON-SOOK LEE. We use multi-resolution image fusion with image laplacian pyramid to merge the boundary.5. A face modeling system based on image, which named Face3DRecon, is designed and implemented. The correctness and high efficiency of the algorithms in this thesis are proven by the system.