3D model based face recognition by face representation using PVM and pose approximation

Abstract

Since a generative 3D face model consists of a large number of vertex points and polygons, a 3D model based face recognition system is generally inefficient in computation time. In this paper, we present a novel 3D face representation method to reduce the number of vertices and optimize its computation time and generate the 3D Korean face model based on the representation method. Also, a pose approximation method is described for initial fitting parameter. Finally, we evaluate the performance of proposed method with the face databases collected using a stereo-camera based 3D face capturing device and a web camera. © Springer-Verlag Berlin Heidelberg 2007.