Toward Bridging the Annotation-Retrieval Gap in Image Search

Ritendra Datta, Weina Ge, Jia Li, and James Z. Wang
The Pennsylvania State University, University Park, PA 16802

We are interested in improving image search quality by bridging the automatic image tagging and retrieval technologies. We argue that for various realworld scenarios of image search, our approach has better potential for producing meaningful search results than image retrieval alone. These cases include single and multi-word search on image collections that have some or no associated tags, and example-based searches, all performed at high speed. To achieve this, we use novel statistical modeling techniques combined with a word ontology. Through extensive experimentation, we show that our method is a promising new approach to image search.

Full Paper in Color
(PDF, 0.8MB)

On-line Info

Citation: Ritendra Datta, Weina Ge, Jia Li and James Z. Wang, ``Toward Bridging the Annotation-Retrieval Gap in Image Search,'' IEEE MultiMedia, vol. 14, no. 3, pp. 24-35, 2007.

Copyright 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works, must be obtained from the IEEE.

Last Modified: April 8, 2007
© 2007