Improving web-image search results using query-relative classifiers - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Improving web-image search results using query-relative classifiers

Résumé

Web image search using text queries has received considerable attention. However, current state-of-the-art approaches require training models for every new query, and are therefore unsuitable for real-world web search applications. The key contribution of this paper is to introduce generic classifiers that are based on query-relative features which can be used for new queries without additional training. They combine textual features, based on the occurence of query terms in web pages and image meta-data, and visual histogram representations of images. The second contribution of the paper is a new database for the evaluation of web image search algorithms. It includes 71478 images returned by a web search engine for 353 different search queries, along with their meta-data and ground-truth annotations. Using this data set, we compared the image ranking performance of our model with that of the search engine, and with an approach that learns a separate classifier for each query. Our generic models that use query-relative features improve significantly over the raw search engine ranking, and also outperform the query-specific models.
Fichier principal
Vignette du fichier
kavj10cvpr.pdf (2.63 Mo) Télécharger le fichier
Vignette du fichier
KAVJ10.png (156.37 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Format Figure, Image
Loading...

Dates et versions

inria-00548636 , version 1 (20-12-2010)

Identifiants

Citer

Josip Krapac, Moray Allan, Jakob Verbeek, Frédéric Jurie. Improving web-image search results using query-relative classifiers. CVPR 2010 - IEEE Conference on Computer Vision & Pattern Recognition, Jun 2010, San Francisco, United States. pp.1094-1101, ⟨10.1109/CVPR.2010.5540092⟩. ⟨inria-00548636⟩
654 Consultations
695 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More