Evaluating dimensionality reduction techniques for visual category recognition using rényi entropy

Gupta, A and Bowden, R (2011) Evaluating dimensionality reduction techniques for visual category recognition using rényi entropy European Signal Processing Conference. pp. 913-917.

	Text Gupta_1569427141.pdf - ["content_typename_Submitted version (pre-print)" not defined] Restricted to Repository staff only Available under License : See the attached licence file. Download (418kB)
	Text (licence) SRI_deposit_agreement.pdf Restricted to Repository staff only Available under License : See the attached licence file. Download (33kB)

Abstract

Visual category recognition is a difficult task of significant interest to the machine learning and vision community. One of the principal hurdles is the high dimensional feature space. This paper evaluates several linear and non-linear dimensionality reduction techniques. A novel evaluation metric, the rényi entropy of the inter-vector euclidean distance distribution, is introduced. This information theoretic measure judges the techniques on their preservation of structure in lower-dimensional sub-space. The popular dataset, Caltech-101 is utilized in the experiments. The results indicate that the techniques which preserve local neighborhood structure performed best amongst the techniques evaluated in this paper. © 2011 EURASIP.

Item Type:	Article
Divisions :	Faculty of Engineering and Physical Sciences > Electronic Engineering > Centre for Vision Speech and Signal Processing
Authors :	Name Email ORCID Gupta, A Bowden, R
Date :	August 2011
Additional Information :	Copyright 2011 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Depositing User :	Symplectic Elements
Date Deposited :	18 Nov 2015 10:03
Last Modified :	31 Oct 2017 17:46
URI:	http://epubs.surrey.ac.uk/id/eprint/808983

Actions (login required)

View Item

Downloads