A single gallery-based face recognition using extended joint sparse representation

For many practical face recognition problems, such as law enforcement, e-passport, ID card identification, and video surveillance, there is usually only a single sample per person enrolled for training, meanwhile the probe samples can usually be captured on the spot, it is possible to collect multiple face images per person. This is a new face recognition problem with many challenges, and we name it as the single-image-to-image-set face recognition problem (ISFR). In this paper, a customized dictionary-based face recognition approach is proposed to solve this problem using the extended joint sparse representation. We first learn a customized variation dictionary from the on-location probing face images, and then propose the extended joint sparse representation, which utilizes the information of both the customized dictionary and the gallery samples, to classify the probe samples. Finally we compare the proposed method with the related methods on several popular face databases, including Yale, AR, CMU-PIE, Georgia, Multi-PIE and LFW databases. The experimental results show that the proposed method outperforms most of these popular face recognition methods for the ISFR problem.