2007 IEEE International Conference on Image Processing - San Antonio, Texas, U.S.A. - September 16-19, 2007

Technical Program

Paper Detail

Paper:MA-L5.2
Session:Biometrics I
Time:Monday, September 17, 10:10 - 10:30
Presentation: Lecture
Title: ORTHOGONAL NEIGHBORHOOD PRESERVING EMBEDDING FOR FACE RECOGNITION
Authors: Xiaoming Liu; Zhejiang University 
 Jianwei Yin; Zhejiang University 
 Zhilin Feng; Zhejiang University of Technology 
 Jinxiang Dong; Zhejiang University 
 Lu Wang; Tsinghua University 
Abstract: In this paper, we propose a new algorithm called Orthogonal Neighborhood Preserving Embedding (ONPE) for face recognition. ONPE can preserve local geometry information and is based on the local linearity assumption that each data point and its k nearest neighbors lie on a linear manifold locally embedded in the image space. ONPE is based on Neighborhood Preserving Embedding (NPE), but overcomes the metric distortion problem of NPE, while metric distortion usually leads to performance degradation. Besides, we propose a classification method (ONPC) based on the ONPE, which use local label propagation method in the reduced space for face recognition. ONPC is based on the natural assumption that the local neighborhood information is also preserved in reduced space, and the label of a data point can be obtained in the reduced space by the labels of its neighbors. Experimental results on two face databases demonstrate the effectiveness of our proposed method.



©2016 Conference Management Services, Inc. -||- email: webmaster@icip2007.com -||- Last updated Friday, August 17, 2012