Determining a suitable metric when using non-negative matrix factorization

The Non-negative Matrix Factorization technique (NMF) has been recently proposed for dimensionality reduction. NMF is able to produce a region- or part-based representation of objects and images. The positive space defined with NMF lacks a suitable metric and this paper experimentally compares NMF t...

Full description

Saved in:
Bibliographic Details
Published inObject recognition supported by user interaction for service robots Vol. 2; pp. 128 - 131 vol.2
Main Authors Guillamet, D., Vitria, J.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2002
Subjects
Computer vision
Covariance matrix
Earth
Eigenvalues and eigenfunctions
Pattern recognition
Principal component analysis
Runtime
Testing
Vectors
Online AccessGet full text

Cover

More Information
Summary:The Non-negative Matrix Factorization technique (NMF) has been recently proposed for dimensionality reduction. NMF is able to produce a region- or part-based representation of objects and images. The positive space defined with NMF lacks a suitable metric and this paper experimentally compares NMF to Principal Component Analysis (PCA) in the context of classification, trying to determine the best distance metric for NMF. This paper introduces the use of the Earth Mover's Distance (EMD) as a relevant metric that takes into account the positive definition of the NMF bases, leading to better recognition results when the dimensionality of the problem is correctly chosen. PCA and NMF have also been tested under the presence of occlusions and, due to its part-based representation, NMF is able to improve on the PCA results.
ISBN:076951695X
ISSN:1051-4651
2831-7475
DOI:10.1109/ICPR.2002.1048254