Principal manifolds and Bayesian subspaces for visual recognition

Author

Moghaddam, Baback

Author_Institution

Mitsubishi Electr. Res. Lab., Cambridge, MA, USA

Volume

2

fYear

1999

fDate

1999

Firstpage

1131

Abstract

We investigate the use of linear and nonlinear principal manifolds for learning low dimensional representations for visual recognition. Three techniques: principal component analysis (PCA), independent component analysis (ICA) and nonlinear PCA (NLPCA) are examined and tested in a visual recognition experiment using a large gallery of facial images from the “FERET” database. We compare the recognition performance of a nearest neighbour matching rule with each principal manifold representation to that of a maximum a posteriori (MAP) matching rule using a Bayesian similarity measure derived from probabilistic subspaces, and demonstrate the superiority of the latter

Keywords

Bayes methods; face recognition; principal component analysis; visual databases; Bayesian similarity measure; Bayesian subspaces; FERET database; ICA; NLPCA; PCA; facial images; independent component analysis; linear principal manifolds; low dimensional representations; maximum a posteriori matching rule; nearest neighbour matching rule; nonlinear PCA; nonlinear principal manifolds; principal component analysis; principal manifold representation; probabilistic subspaces; recognition performance; visual recognition; Bayesian methods; Ear; Face recognition; Image analysis; Image recognition; Independent component analysis; Karhunen-Loeve transforms; Laboratories; Principal component analysis; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Vision, 1999. The Proceedings of the Seventh IEEE International Conference on

Conference_Location

Kerkyra

Print_ISBN

0-7695-0164-8

Type

conf

DOI

10.1109/ICCV.1999.790407

Filename

790407