Face recognition via compact second order image gradient orientations

TL;DR

This paper introduces a novel face recognition method using compact second order image gradient orientations (CSOIGO), which captures richer geometric information and improves robustness against noise and occlusion, outperforming existing approaches.

Contribution

The paper proposes a new compact second order image gradient orientation (CSOIGO) method using linear complex PCA, enhancing face recognition accuracy especially with limited training data.

Findings

CSOIGO outperforms competing methods with few training samples.

The approach is robust against disguise, occlusion, and mixed variations.

It surpasses some deep neural network approaches in accuracy.

Abstract

Conventional subspace learning approaches based on image gradient orientations only employ the first-order gradient information. However, recent researches on human vision system (HVS) uncover that the neural image is a landscape or a surface whose geometric properties can be captured through the second order gradient information. The second order image gradient orientations (SOIGO) can mitigate the adverse effect of noises in face images. To reduce the redundancy of SOIGO, we propose compact SOIGO (CSOIGO) by applying linear complex principal component analysis (PCA) in SOIGO. Combined with collaborative representation based classification (CRC) algorithm, the classification performance of CSOIGO is further enhanced. CSOIGO is evaluated under real-world disguise, synthesized occlusion and mixed variations. Experimental results indicate that the proposed method is superior to its competing approaches with few training samples, and even outperforms some prevailing deep neural network based approaches. The source code of CSOIGO is available at https://github.com/yinhefeng/SOIGO.