Face Recognition System based on Face Pose Estimation and Frontal Face Pose Synthesis

Face Recognition (FR) has a wide range of applications, such as face-based video indexing, human-computer interaction, and surveillance system. The recognition rate can reach more than 90% with the frontal face, while dealing with faces are oriented or rotated, the accuracy just achieves 40% or 50%. Hence, an algorithm is proposed in this paper to deal with the problem. The face pose is first determined and then a frontal face pose is synthesized by wire frame model (WFM) according to the face pose parameter. In the FR system, Gabor filters are adopted for feature extraction, while the subspace based algorithms, principal component analysis (PCA) and linear discriminant analysis (LDA), are used for dimension reduction. The 2-norm distance is applied for similarity at last. With synthesized frontal face pose information, it can be seen from experimental results that the recognition rate is greatly improved. Keywords-face recognition; wire frame model; principal component analysis; linear discriminant analysis;