Search

The single sample per person (SSPP) problem is of great importance for real-world face recognition systems. In SSPP scenario, there is always a large gap between a normal sample enrolled in the gallery set and the non-ideal probe sample. It is a crucial step for face recognition with SSPP to bridge the gap between the ideal and non-ideal samples. For this purpose, we propose a Variational Feature Representation-based Classification (VFRC) method, which employs the linear regression model to fit the variational information of a non-ideal probe sample with respect to an ideal gallery sample. Thus, a corresponding normal feature, which reserve the identity information of the probe sample, is obtained. A combination of the normal feature and the probe sample is used, which makes VFRC method more robust and effective for SSPP scenario. The experimental results show that VFRC method possesses higher recognition rate than other related face recognition methods. [ABSTRACT FROM AUTHOR]

In the field of face recognition, the small sample size (SSS) problem and non‐ideal situations of facial images are recognised as two of the most challenging issues. Recently, Zhu et al. proposed a patch‐based collaborative representation (PCRC) method which showed good performance for the SSS and the single sample per person problems; and Peng et al. proposed a locality‐constrained collaborative representation (LCCR) method which achieved high robustness for face recognition in non‐ideal situations. Inspired by the methods proposed in PCRC and LCCR, this study proposes a patch‐based locality‐enhanced collaborative representation (PLECR) method to combine and enhance the advantages of both PCRC and LCCR. The PLECR and several related methods are implemented on AR, face recognition technology and extended Yale B databases; and the extensive numerical results show that PLECR is more efficient among these methods for the SSS problem in non‐ideal situations, especially for the SSS problem with occlusions. [ABSTRACT FROM AUTHOR]