Karhunen-Loéve expansion of a set of rotated templates

Matjaz Jogan, Emil Zagar, Ales Leonardis

Abstract

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.
BibTeX
@article{jogan2003karhunen,
  author = {Matjaz Jogan and Emil Zagar and Ales Leonardis},
  title = {{Karhunen-Loéve expansion of a set of rotated templates}},
  journal = {IEEE Transactions on Image Processing},
  year = {2003},
  doi = {10.1109/tip.2003.813141},
}

Loading related content...

Figure