Two-view face recognition using Bayesian fusion

Two-dimensional face recognition suffered from pose changes, while three-dimensional approaches are with high computational complexity. Motivated by this, a two-view face recognition system for digital home is presented in this paper. Besides the improvement in recognition rate, this system reduces the misclassification that could occur in traditional single-view systems. The proposed system fuses the individual recognition results of two images of the same identity with different viewing angles based on Bayesian theory. Bayesian approach uses the similarity of each person and is trained by determining the reliability of each identity of the two channels. A frontal view and a side view are chosen since they convey the most important information of human faces. Each input image is sent into its corresponding channel to obtain a 2D face recognition result. Within each channel, PCA and SVM are applied. Different form traditional PCA based approaches, SVM classifiers are used instead of minimum distance classifier to enhance the robustness. Our experimental results show that this two-view face recognition system has achieved a higher recognition rate compared with traditional 2D single-view face recognition systems.