Single sample face recognition based on multiple features and twice classification

Authors:
Xiaohua Wang;Wei Liu;Min Hu;Liangfeng Xu
Affiliations:
School of Computer and Information, Hefei University of Technology, China,Affective Computing and Advanced Intelligent Machines Anhui Key Laboratory, China;School of Computer and Information, Hefei University of Technology, China,Affective Computing and Advanced Intelligent Machines Anhui Key Laboratory, China;School of Computer and Information, Hefei University of Technology, China,Affective Computing and Advanced Intelligent Machines Anhui Key Laboratory, China;School of Computer and Information, Hefei University of Technology, China
Venue:
ICIC'13 Proceedings of the 9th international conference on Intelligent Computing Theories
Year:
2013

Citing 6
Cited 0

Distortion Invariant Object Recognition in the Dynamic Link Architecture

IEEE Transactions on Computers
Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns

IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic Texture Synthesis for Face Recognition from Single Views

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 03
Face Description with Local Binary Patterns: Application to Face Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
WLD: A Robust Local Image Descriptor

IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Recognition with Single Training Sample Based on Local Feature Fusion

DCABES '11 Proceedings of the 2011 10th International Symposium on Distributed Computing and Applications to Business, Engineering and Science

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to improve the performance of face recognition with single sample effectively, a face recognition method based on multiple features and twice classification is proposed. For obtaining sufficient face information, facial multiple features combining differential excitation and Compound Local Binary Pattern (CLBP) on the asymmetric region are extracted. Elastic Matching (EM) has better robustness for pose. However, the computation complexity of the method is rather high. Classifying twice strategy is proposed to short the time of data processing. Experimental results on ORL database and FERET database show that the method is effective in getting better recognition rate and speed, also has a certain robustness to pose.