SIFT-PCA: An efficient and accurate face representation and recognition model

Abstract

In this paper, a new technique called SIFT-PCA is devised for efficient and accurate face representation and recognition. Discriminative SIFT descriptors are extracted for compact representation. Unlike PCA-SIFT that employs PCA on normalized gradient patch, we employ PCA on smoothed weighted histograms. The proposed model has better computing performance in terms of recognition rate than the basic SIFT model. To establish the superiority of the proposed model, we have experimentally compared the performance of our new algorithm with recently proposed (2D)2-PCA on the benchmark databases: AT&T and UMIST face datasets. Copyright © 2009 by IICAI.