Multi view video based face recognition

With the prevalence of camera networks, multi view surveillance video have become commons. Multi view face recognition has become an active research area in recent years. In this paper, an approach for video-based face recognition in camera networks is proposed. Video scenes have unlimited orientation and poses. Video provide an efficient way for feature extraction. The proposed feature is developed using the spherical harmonic representation of the face. The texture map of the face is constructed by projecting the image intensity values on the surface of the spherical model. A particle filter is used to track the 3D location of the face. The similarity between feature sets can be measured using the reproducing Kernel Hilbert space.