User experiments with the Eurovision cross-language image retrieval system

In this paper we present Eurovision, a text-based system for cross-language (CL) image retrieval. The system is evaluated by multilingual users for two search tasks with the system configured in English and five other languages. To our knowledge this is the first published set of user experiments for CL image retrieval. We show that: (1) it is possible to create a usable multilingual search engine using little knowledge of any language other than English, (2) categorizing images assists the user's search, and (3) there are differences in the way users search between the proposed search tasks. Based on the two search tasks and user feedback, we describe important aspects of any CL image retrieval system.