A SVM face recognition method based on Gabor-featured key points

• Published in: 2005 International Conference on Machine Learning and Cybernetics Vol. 8; pp. 5144 - 5149 Vol. 8
• Main Authors: Jun Qin, Zhong-Shi He
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2005
• Subjects: Face recognition; Frequency; Gabor Feature; Humans; Kernel; Linear discriminant analysis; Machine learning; Robustness; Spatial databases; Support Vector Machine; Support vector machine classification; Support vector machines
• ISBN: 0780390911; 9780780390911
• ISSN: 2160-133X
• DOI: 10.1109/ICMLC.2005.1527850

This paper presents a novel face recognition approach based on support vector machine and Gabor-featured key points, which takes technological advantages of both support vector machine and Gabor feature extraction. The main contributions of this paper therefore lie in the following aspects: (1) support vector machine is successfully applied to face recognition by using Gabor features of key points; (2) Gabor features of key points are introduced to represent a whole face in a computable dimensional space. As a result, experiments on FERET and AT&T databases have shown significant better performance with this method, which in itself proves the feasibility of our proposal.