Bimode Mode for Face Recognition

Tensorface based approaches decompose an image into its constituent factors (i.e. person, lighting, viewpoint, etc.) and then utilize these factor spaces for recognition. However, tensorface is not a preferable choice because of the complexity of its multimode. In addition, a single mode space, except the person space, could not be used for recognition directly. From the viewpoint of practical application, we propose a bimode mode for face recognition. This new mode can be treated as a simplified mode representation of tensorface. However, their respective algorithms for training are completely different due to their different definitions of subspaces. Thanks to its simpler mode form, the proposed mode requires less iteration times in the process of training and testing. Comprehensive experiments on three face image databases (PEAL, YaleB frontal and Weizmann) validate the effectiveness of the proposed new mode.