Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

High-dimensional visual vocabularies for image retrieval

High-dimensional visual vocabularies for image retrieval SIGIR 2007 Proceedings Poster High-Dimensional Visual Vocabularies for Image Retrieval João Magalhães1, Stefan Rüger1,2 Department of Computing Imperial College London South Kensington Campus London SW7 2AZ, UK Knowledge Media Institute The Open University Walton Hall Milton Keynes MK7 6AA, UK (j.magalhaes@imperial.ac.uk, s.rueger@open.ac.uk) ABSTRACT In this paper we formulate image retrieval by text query as a vector space classification problem. This is achieved by creating a high-dimensional visual vocabulary that represents the image documents in great detail. We show how the representation of these image documents enables the application of well known text retrieval techniques such as Rocchio tf-idf and na ve Bayes to the semantic image retrieval problem. We tested these methods on a Corel images subset and achieve state-of-the-art retrieval performance using the proposed methods. characteristics (colour and texture features). Since we are going to use a single vocabulary to represent all images, we need a set of visual terms that is able to represent them. Thus, we need to check which visual characteristics are more common in the dataset. For example, if there are a lot of images with a wide range of blue tones we require a larger number of visual terms representing the different http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

High-dimensional visual vocabularies for image retrieval

Association for Computing Machinery — Jul 23, 2007

Loading next page...
/lp/association-for-computing-machinery/high-dimensional-visual-vocabularies-for-image-retrieval-JVE5l9tLfW

References (8)

Datasource
Association for Computing Machinery
Copyright
Copyright © 2007 by ACM Inc.
ISBN
978-1-59593-597-7
doi
10.1145/1277741.1277923
Publisher site
See Article on Publisher Site

Abstract

SIGIR 2007 Proceedings Poster High-Dimensional Visual Vocabularies for Image Retrieval João Magalhães1, Stefan Rüger1,2 Department of Computing Imperial College London South Kensington Campus London SW7 2AZ, UK Knowledge Media Institute The Open University Walton Hall Milton Keynes MK7 6AA, UK (j.magalhaes@imperial.ac.uk, s.rueger@open.ac.uk) ABSTRACT In this paper we formulate image retrieval by text query as a vector space classification problem. This is achieved by creating a high-dimensional visual vocabulary that represents the image documents in great detail. We show how the representation of these image documents enables the application of well known text retrieval techniques such as Rocchio tf-idf and na ve Bayes to the semantic image retrieval problem. We tested these methods on a Corel images subset and achieve state-of-the-art retrieval performance using the proposed methods. characteristics (colour and texture features). Since we are going to use a single vocabulary to represent all images, we need a set of visual terms that is able to represent them. Thus, we need to check which visual characteristics are more common in the dataset. For example, if there are a lot of images with a wide range of blue tones we require a larger number of visual terms representing the different

There are no references for this article.