From Text to Image: Generating Visual Query for Image Retrieval

In this paper, we explore the help of visual features to cross-language image retrieval. We propose an approach that transforms textual queries into visual representations. The relationships between text and images are modeled. Visual queries are constructed from textual queries using the relationships. The retrieval results using textual and visual queries are combined to generate the final ranked list. We conducted English monolingual and Chinese-English cross-language retrieval experiments. The performances are quite good. The average precision of English monolingual textual run is 0.6304. The performance of cross-lingual retrieval is about 70% of monolingual retrieval. However, the help of generated visual query is limit. If appropriate query terms are selected to generate visual query, retrieval performance could be increased.

• Type: Conference
• Year: 2004
• Where: CLEF
• Authors: Wen-Cheng Lin, Yih-Chen Chang, Hsin-Hsi Chen
• Keywords: CLEF 2004, Textual Queries, Visual, Visual Queries