Share Email Print

Journal of Electronic Imaging

Improved similarity measure-based graph embedding for face recognition
Author(s): Yongxin Ge; Dan Yang; Xiaohong Zhang; Jiwen Lu
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

We propose an improved similarity measure (ISM) and apply it to the existing graph embedding (GE) framework to derive a new improved similarity measure-based graph embedding (ISM-GE) method for face recognition. Our work is motivated by the fact that both the Euclidean metric and the correlation metric are useful and effective for characterizing the similarity of face samples, and we combine these two metrics to form a new ISM to measure the similarity of face samples. We further utilize the proposed ISM in the existing GE framework and develop a new ISM-GE method for face feature extraction and recognition. Experimental results on two widely used face databases demonstrate the efficacy of the proposed method.

Paper Details

Date Published: 22 February 2012
PDF: 8 pages
J. Electron. Imag. 21(1) 013002 doi: 10.1117/1.JEI.21.1.013002
Published in: Journal of Electronic Imaging Volume 21, Issue 1
Show Author Affiliations
Yongxin Ge, Chongqing Univ. (China)
Dan Yang, Chongqing Univ. (China)
Xiaohong Zhang, Chongqing Univ. (China)
Jiwen Lu, Advanced Digital Sciences Ctr. (Singapore)

© SPIE. Terms of Use
Back to Top