3D Face Modeling

Generating realistic 3D human face models has been a persistent challenger in computer graphics. With the development of virtual reality, movie making, visual telephone, etc., this research area is now receiving more attention than ever before. This thesis makes a deeply research on the area.This thesis gives the future application of 3D face models and the present research; defines the facial feature points on the research of human face's character; researches on extracting facial feature points from human face's images, and gives the detail algorithm; finishes generating 3D face model using the defined facial feature points. The main research topics include:1) Defining facial feature points. The topic is to define the feature points of a real face. These facial feature points must show not only the whole outline of a face, but also the local character of the face components. On the research of human cranium and head muscle, the definition of the facial feature points is given.2) Image based facial feature points extraction. The topic is to extract facial feature points from face images given. We use the theory of computer stereo vision and the method of distortion on template of facial feature points to extract data from face images.3) 3D face modeling. The content of the research is to generate specific 3D face model using the facial feature points. We finished synthesizing 3D face model of the specific person by modifying the common face model using the person's facial feature points.As the research and application of 3D face modeling, Face3Drecon is developed using Visual C++6.0. The system can be used to analyze the feature of face, extract face feature points from front and side face images, generate 3D face model from face feature points. For generating 3D face models from front and side images, the result looks realistic.