We present a novel method of face verification which is based on the concept and principles of sparse representation of signals. The sparse representation techniques are used in both feature extraction and classification steps. The proposed method is relatively invariant to changes in imaging conditions such as illumination variations. This is due to the characteristics of the sparse sampling method. In order to improve the invariance properties of the system, the feature extraction algorithm is motivated by using the Local Binary Pattern (LBP) features. Our experimental studies on the XM2VTS and XM2VTS-DARK datasets demonstrate that the proposed method improves the performance of the verification system.
Rights and permissions | |
![]() | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |