Karhunen-Loéve expansion of a set of rotated templates

Matjaz Jogan; Emil Zagar; Ales Leonardis

doi:10.1109/TIP.2003.813141

Karhunen-Loéve expansion of a set of rotated templates

IEEE Trans Image Process. 2003;12(7):817-25. doi: 10.1109/TIP.2003.813141.

Authors

Matjaz Jogan¹, Emil Zagar, Ales Leonardis

Affiliation

¹ Faculty of Computer and Information Science, University of Ljubljana, SI-1001 Ljubljana, Slovenia. matjaz.jogan@fri.uni-lj.si

PMID: 18237956
DOI: 10.1109/TIP.2003.813141

Abstract

In this paper, we propose a novel method for efficiently calculating the eigenvectors of uniformly rotated images of a set of templates. As we show, the images can be optimally approximated by a linear series of eigenvectors which can be calculated without actually decomposing the sample covariance matrix.