|
|||||
|
|
Sparse representation for face recognition by discriminative low-rank matrix recovery |
|
| Affiliation: | 1. Department of Animal Science and Biotechnology, Tunghai University, No. 1727, Section 4, Taiwan Boulevard, Taichung City 40704, Taiwan;2. Department of Electrical Engineering, National Yunlin University of Science and Technology, No. 123, University Road, Section 3, Douliou, Yunlin 64002, Taiwan;1. Beijing Key Lab of Traffic Data Analysis and Mining, School of Computer and Information Technology, Beijing Jiaotong University, Beijing, China;2. Key Lab of Machine Learning and Computational Intelligence, College of Mathematics and Computer Science, Hebei University, Hebei, China;1. Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian, China;2. School of Software Technology, Dalian University of Technology, Dalian, China;3. Key Laboratory of Machine Perception (MOE), School of EECS, Peking University, Beijing, China;4. School of Mathematical Sciences, Dalian University of Technology, Dalian, China;5. School of Computing and Mathematics, Charles Sturt University, Bathurst, NSW 2795, Australia;1. No.38 Research Institute, China Electronic Technology Group Corporation, Hefei 230088, PR China;2. School of Computer Science and Technology, Nanjing University of Science and Technology, Nanjing 210094, PR China |
| Abstract: | This paper proposes a discriminative low-rank representation (DLRR) method for face recognition in which both the training and test samples are corrupted owing to variations in occlusion and disguise. The proposed method extends the sparse representation-based classification algorithm by incorporating the low-rank structure of data representation. The DLRR algorithm recovers a clean dictionary with enhanced discrimination ability from the corrupted training samples for sparse representation. Simultaneously, it learns a low-rank projection matrix to correct corrupted test samples by projecting them onto their corresponding underlying subspaces. The dictionary elements from different classes are encouraged to be as independent as possible by regularizing the structural incoherence of the original training samples. This leads to a compact representation of a corrected test sample by a linear combination of more dictionary elements from the corrected class. The experimental results on benchmark databases show the effectiveness and robustness of our face recognition technique. |
| Keywords: | Sparse representation Low-rank representation Matrix recovery Dictionary learning Face recognition A low-rank projection matrix Subspace Eigenface |
| 本文献已被 ScienceDirect 等数据库收录！ | |