Image Sense Classification in Text-Based Image Retrieval

Authors:
Yih-Chen Chang;Hsin-Hsi Chen
Affiliations:
Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan;Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan
Venue:
AIRS '09 Proceedings of the 5th Asia Information Retrieval Symposium on Information Retrieval Technology
Year:
2009

Citing 5
Cited 1

Unsupervised word sense disambiguation rivaling supervised methods

ACL '95 Proceedings of the 33rd annual meeting on Association for Computational Linguistics
Hierarchical clustering of WWW image search results using visual, textual and link information

Proceedings of the 12th annual ACM international conference on Multimedia
Discriminating image senses by clustering with multimodal features

COLING-ACL '06 Proceedings of the COLING/ACL on Main conference poster sessions
Toward a common semantics between media and languages

Proceedings of the 2006 international workshop on Research issues in digital libraries
Approaches of using a word-image ontology and an annotated image corpus as intermedia for cross-language image retrieval

CLEF'06 Proceedings of the 7th international conference on Cross-Language Evaluation Forum: evaluation of multilingual and multi-modal information retrieval

Joint image and word sense discrimination for image retrieval

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

An image sense is a graphic representation of a concept denoted by a (set of) term(s). This paper proposes algorithms to find image senses for a concept, collect the sense descriptions, and employ them to disambiguate the image senses in text-based image retrieval. In the experiments on 10 ambiguous terms, 97.12% of image senses returned by a search engine are covered. The average precision of sample images is 68.26%. We propose four kinds of classifiers using text, image, URL, and expanded text features, respectively, and a merge strategy to combine the results of these classifiers. The merge classifier achieves 0.3974 in F-measure (β=0.5), which is much better than the baseline and has 51.61% of human performance.