DMCA
Separable Linear Discriminant Classification (2005)
| Citations: | 2 - 1 self |
BibTeX
@MISC{Bauckhage05separablelinear,
author = {Christian Bauckhage and John K. Tsotsos},
title = {Separable Linear Discriminant Classification},
year = {2005}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Linear discriminant analysis is a popular technique in computer vision, machine learning and data mining. It has been successfully applied to various problems, and there are numerous variations of the original approach. This paper introduces the idea of separable LDA. Towards the problem of binary classification for visual object recognition, we derive an algorithm for training separable discriminant classifiers. Our approach provides rapid training and runtime behavior and also tackles the small sample size problem. Experimental results show that the method performs robust and allows for online learning.
Keyphrases
separable linear discriminant classification binary classification popular technique rapid training linear discriminant analysis computer vision machine learning various problem online learning visual object recognition data mining small sample size problem numerous variation runtime behavior original approach experimental result separable discriminant classifier separable lda
