Abstract:Sparse representation based classification (SRC) has recently been proposed for robust face recognition. To deal with occlusion, SRC introduces an identity matrix as an occlusion dictionary on the assumption that the occlusion has sparse representation in this dictionary. However, the results show that SRC's use of this occlusion dictionary is not nearly as robust to large occlusion as it is to random pixel corruption. In addition, the identity matrix renders the expanded dictionary large, which results in expensive computation. In this paper, we present a novel method, namely structured sparse representation based classification (SSRC), for face recognition with occlusion. A novel structured dictionary learning method is proposed to learn an occlusion dictionary from the data instead of an identity matrix. Specifically, a mutual incoherence of dictionaries regularization term is incorporated into the dictionary learning objective function which encourages the occlusion dictionary to be as independent as possible of the training sample dictionary. So that the occlusion can then be sparsely represented by the linear combination of the atoms from the learned occlusion dictionary and effectively separated from the occluded face image. The classification can thus be efficiently carried out on the recovered non-occluded face images and the size of the expanded dictionary is also much smaller than that used in SRC. The extensive experiments demonstrate that the proposed method achieves better results than the existing sparse representation based face recognition methods, especially in dealing with large region contiguous occlusion and severe illumination variation, while the computational cost is much lower.

Robust Face Recognition under Varying Illumination and Occlusion Considering Structured Sparsity

Robust Face Recognition by Fusion Local Singular Value Feature and Deformable Model

Local Spatial Continuity Steered Sparse Representation for Occluded Face Recognition

A Fusion Framework For Face Recognition Under Varying Illumination Based On Multi-Scale Analysis

Face recognition with contiguous occlusion using linear regression and level set method

Robust Face Recognition with Occlusions in Both Reference and Query Images

Robust Face Recognition Via Occlusion Dictionary Learning.

Robust and Practical Face Recognition Via Structured Sparsity

Robust Face Recognition Via Sparse Representation

Structured Occlusion Coding for Robust Face Recognition

Extracting sparse error of robust PCA for face recognition in the presence of varying illumination and occlusion

Robust Face Recognition with Locality-Sensitive Sparsity and Group Sparsity Constraints.

Robust Matrix Regression for Illumination and Occlusion Tolerant Face Recognition

Structured Sparse Error Coding for Face Recognition with Occlusion

Adaptive Illumination-Invariant Face Recognition Via Local Nonlinear Multi-Layer Contrast Feature.

Robust Face Image Alignment Using Structural Priors

Robust Face Recognition with Kernelized Locality-Sensitive Group Sparsity Representation

Learning a Sparse Representation for Robust Face Recognition.

Robust Low-Rank Regularized Regression for Face Recognition with Occlusion

A Robust Face Recognition Method Combining LBP with Multi-mirror Symmetry for Images with Various Face Interferences.

Robust Face Recognition by Sparse Local Features from a Single Image under Occlusion