Image annotation refinement using Web-based keyword correlation

Llorente, Ainhoa; Motta, Enrico and Rüger, Stefan (2009). Image annotation refinement using Web-based keyword correlation. In: Proceedings of the 4th International Conference on Semantic and Digital Media Technologies, 2-4 Dec 2009, Graz, Austria.

Full text available as:

PDF (Version of Record) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (97Kb) | Request Copy from OU Author

DOI (Digital Object Identifier) Link:	http://dx.doi.org/doi:10.1007/978-3-642-10543-2_22
Google Scholar	Look up in Google Scholar

Abstract

This paper describes a novel approach that automatically refines the image annotations generated by a non-parametric density estimation model. We re-rank these initial annotations following a heuristic algorithm, which uses semantic relatedness measures based on keyword correlation on the Web. Existing approaches that rely on keyword co-occurrence can exhibit limitations, as their performance depend on the quality and coverage provided by the training data. Additionally, WordNet based correlation approaches are not able to cope with words that are not in the thesaurus. We illustrate the effectiveness of our Web-based approach by showing some promising results obtained on two datasets, Corel 5k, and ImageCLEF2009.

Item Type:	Conference or Workshop Item
Copyright Holders:	2009 Springer-Verlag Berlin Heidelberg
ISSN:	0302-9743
Keywords:	automated image annotation; normalized Google distance; semantic similarity
Academic Unit/Department:	Knowledge Media Institute
Interdisciplinary Research Centre:	Centre for Research in Computing (CRC)
Item ID:	23440
Depositing User:	Kay Dave
Date Deposited:	14 Oct 2010 12:33
Last Modified:	07 Apr 2011 10:17
URI:	http://oro.open.ac.uk/id/eprint/23440

	Repository Staff Only: edit this item
Public: Report issue/request change

Policies | Disclaimer