DocumentCode :
2828993
Title :
Discriminant subclass-center manifold preserving projection for face feature extraction
Author :
Lan, Chao ; Jing, Xiaoyuan ; Zhang, David ; Gao, Shiqiang ; Yang, Jingyu
Author_Institution :
State Key Lab. for Software Eng., Wuhan Univ., Wuhan, China
fYear :
2011
fDate :
11-14 Sept. 2011
Firstpage :
3013
Lastpage :
3016
Abstract :
Manifold learning is an effective feature extraction technique, which seeks a low-dimensional space where the manifold structure, in terms of local neighborhood, of the data set can be well preserved. A typical manifold learning method constructs a local neighborhood centered at individual samples. In this paper, we propose to construct local neighborhoods that centered at subclass centers, and seek an embedded space where such neighborhood is well preserved. We show from a probability perspective that, neighbors of a subclass center would contain more intra-class data than inter-class data, which may be desirable for discrimination. Meanwhile, we simultaneously enhance the discriminative power of extracted features by maximizing the Fisher ratio of embedded data based on subclass centers. Experimental results on CAS-PEAL and FERET face databases demonstrate that our proposed approach is more effective than most typical manifold learning methods and their supervised extensions in classification performance.
Keywords :
face recognition; feature extraction; learning (artificial intelligence); visual databases; FERET face databases; Fisher ratio; discriminant subclass center manifold preserving projection; embedded space; face feature extraction; manifold learning; manifold learning method; manifold structure; Databases; Face; Face recognition; Feature extraction; Learning systems; Manifolds; Training; Discriminant subclass-center manifold preserving projection (DSMPP); Face feature extraction; Manifold learning; Subclass-center neighborhood structure;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing (ICIP), 2011 18th IEEE International Conference on
Conference_Location :
Brussels
ISSN :
1522-4880
Print_ISBN :
978-1-4577-1304-0
Electronic_ISBN :
1522-4880
Type :
conf
DOI :
10.1109/ICIP.2011.6116297
Filename :
6116297
Link To Document :
بازگشت