3D reconstruction is a hot spot in the field of computer vision. Based on a depth sensor characteristics of real time and low cost of Kinect Fusion algorithm

a method which can automatically recognize face region by obtaining depth data and RGB data to reconstruct the 3D model on the basis of Kinect Fusion. This method adds a penalty factor to smooth the motivation of human face. Texture integration is applied to get a color information.GPU acceleration is applied during the reconstruction.