Structure-constrained Discriminative Dictionary Learning Based on Schatten P-Norm for Face Recognition

Heyou Chang,Fanlong Zhang,Guangwei Gao,Hao Zheng,Yang Chen
DOI: https://doi.org/10.1016/j.dsp.2019.08.003
IF: 2.92
2019-01-01
Digital Signal Processing
Abstract:Dictionary learning based on low-rank representations has been widely studied in recent years and has achieved impressive performance in face recognition. However, studies on the case where both the training and testing data are polluted with large noise are still in their infancy. In this paper, to capture the essential structure hidden in noisy data and to learn a discriminative dictionary, we propose a novel model of structure-constrained discriminative dictionary learning based on the Schatten p -norm (SDDLS p ), in which (1) the nonconvex Schatten p -norm (0 < p < 1) is used to approximate the rank minimization problem, (2) a weighted representation regularization term is constructed to generate a coefficient matrix with a class-wise block-diagonal structure, and (3) a linear classifier based on the coefficient is simultaneously optimized with the dictionary and representation update. Thus, the learned dictionary and representation perform well in face recognition problems, particularly with training and testing images that are both polluted by large noises. An iterative algorithm to solve the proposed SDDLS p is also presented in this paper. Experiments on four face databases demonstrate the advantages of the proposed method over previous dictionary-learning methods.
What problem does this paper attempt to address?