A Gabor Feature Classifier for Face Recognition

This paper describes a novel Gabor Feature Class$er (GFC)method forface recognition. The GFC method employs an enhanced Fisher discrimination model on an augmented Gabor feature vector; which is derived from the Gabor wavelet transformation office images. The Gabor wavelets, whose kernels are similar to the 2 0 receptivefield profiles of the nianinialian cortical simple cells, exhibit desirable characteristics of spatial locality and orientationselectivity. As a result, the Gabor transformed face images produce salient local and discriminating features that are suitableforface recognition. Thefeasibility of the new GFC method has been successfully tested onface recognition using 600 FERET frontal face images, which involve different illumination and varied facial expressions of 200 subjects. The effectiveness of the novel GFC method is shown in ternis of both absolute performance indices and comparative performance against some popular face recognition schemes such as the Eigenfaces m...

• Type: Conference
• Year: 2001
• Where: ICCV
• Authors: Chengjun Liu, Harry Wechsler
• Keywords: Augmented Gabor Feature, Computer Vision, Face Recognition Schemes, Gabor Transformed Face, Gabor Wavelet Transformation, Gabor Wavelets, ICCV 2001