CWI used PF/Tijah, a flexible XML retrieval system, to evaluate image retrieval based on textual evidence in the context of the wikipediaMM task at ImageCLEF 2008. We employed a language modelling framework and found that the text associated with the Wikipedia images is a good source of evidence. We also investigated a length prior and found that biasing towards images with longer descriptions than the ones retrieved by our language modelling approach is not benficial.

,
CLEF-campaign
Image Indexing and reTrievAL in the Large Scale
Workshop on Cross-Language Evaluation Forum
Human-Centered Data Analytics

Tsikrika, T., Rode, H., & de Vries, A. (2008). CWI at ImageCLEF 2008. In Working notes of the 9th Workshop of the Cross-Language Evaluation Forum. CLEF-campaign.