GCS Kernel For SVM-Based Image Recognition
Abstract
In this paper, we present a new compactly supported kernel for SVM based image recognition. This kernel which we called Geometric Compactly Supported (GCS) can be viewed as a generalization of spherical kernels to higher dimensions. The construction of the GCS kernel is based on a geometric approach using the intersection volume of two n-dimensional balls. The compactness property of the GCS kernel leads to a sparse Gram matrix which enhances computation efficiency by using sparse linear algebra algorithms. Comparisons of the GCS kernel performance, for image recognition task, with other known kernels prove the interest of this new kernel.
Reference
@inproceedings{jpt-icann05b,
author = {Boughorbel, S. and Tarel, J.-P. and Fleuret, F. and Boujemaa, N.},
title = {GCS Kernel For SVM-Based Image Recognition},
booktitle = {Proceedings of International Conference on Artificial Neural Networks (ICANN'05)},
date = {September 11-15},
address = {Warsaw, Poland},
volume = {II},
pages = {595 - 600},
year = {2005},
note = {http://perso.lcpc.fr/tarel.jean-philippe/publis/icann05b.html}
}
Pdf file (12169 Kb)
(c) Springer-Verlag