Three-dimensional reconstruction of face model based on single photo

Creating a realistic three-dimensional human face model has been one of the most difficult problems in the fields of computer graphics and computer vision. A method witch to build three-dimensional human face model basic on a photo of two-dimensional single front face is proposed in this paper. Firstly, facial organ distribution is identified from the photo by face recognition. After that, the standard three-dimensional human face model is modified to match the face on the photo. At the same time, feature triangle mesh which consists thousands of feature points would be created by triangular decomposition of the photo. The value of X and Y coordinates of those feature points are identified. In the next step, the feature points are substituted into the modified face model and the value of Z coordinate of each feature points will be calculated. Finally, the 3D human face model is built by all those feature points. And the created 3D face model will be rendered and shown by OpenGL. The method we discussed is simple and rapid. The input is just a single frontal photo. It´s proved that this method has good results.