Equivalence of Some Common Linear Feature Extraction Techniques for Appearance-Based Object Recognition Tasks

Author

Asunción Vicente, M. ; Hoyer, Patrik O. ; Hyvärinen, Aapo

Author_Institution

Dept. of Ind. Syst. Eng., Miguel Hernandez Univ., Alicante

Volume

29

Issue

5

fYear

2007

fDate

5/1/2007 12:00:00 AM

Firstpage

896

Lastpage

900

Abstract

Recently, a number of empirical studies have compared the performance of PCA and ICA as feature extraction methods in appearance-based object recognition systems, with mixed and seemingly contradictory results. In this paper, we briefly describe the connection between the two methods and argue that whitened PCA may yield identical results to ICA in some cases. Furthermore, we describe the specific situations in which ICA might significantly improve on PCA

Keywords

feature extraction; independent component analysis; object recognition; principal component analysis; appearance-based object recognition tasks; computer vision; independent component analysis; linear feature extraction techniques; principal component analysis; Covariance matrix; Eigenvalues and eigenfunctions; Feature extraction; Independent component analysis; Layout; Lighting; Object recognition; Principal component analysis; Reflectivity; Shape; Computer vision; independent component analysis.; object recognition; principal component analysis; Algorithms; Artificial Intelligence; Image Enhancement; Image Interpretation, Computer-Assisted; Imaging, Three-Dimensional; Linear Models; Models, Biological; Models, Statistical; Pattern Recognition, Automated; Principal Component Analysis; Reproducibility of Results; Sensitivity and Specificity;

fLanguage

English

Journal_Title

Pattern Analysis and Machine Intelligence, IEEE Transactions on

Publisher

ieee

ISSN

0162-8828

Type

jour

DOI

10.1109/TPAMI.2007.1074

Filename

4135683