An Unsupervised Learning Approach to Content-based Image Retrieval

Yixin Chen, James Z. Wang
The Pennsylvania State University, University Park, PA 16802

Robert Krovetz
NEC Corporation
Abstract:

''Semantic gap'' is an open challenging problem in content-based image retrieval. It reflects the discrepancy between low-level imagery features used by the ret rieval algorithm and high-level concepts required by system users. This paper in troduces a novel image retrieval scheme, CLUster-based rEtrieval of images by un supervised learning (CLUE), to tackle the semantic gap problem. CLUE is built on a hypothesis that {\it images of the same semantics tend to be clustered}. It a ttempts to narrow the semantic gap by retrieving image clusters based on not onl y the feature similarity of images to the query, but also how images are similar to each other. CLUE has been tested using examples from a database of about $60,000$ general-purpose images. Empirical results demonstrate the effectiveness of CLUE.

Full Paper in Color
(PDF, 0.7MB)

On-line Info

Citation: Yixin Chen, James Z. Wang and Robert Krovetz, An Unsupervised Learning Approach to Content-Based Image Retrieval,'' Proc. IEEE International Symposium on Signal Processing and its Applications, vol. 1, pp. 197-200, Paris, France, July 2003.

Copyright 2003 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.