Translator Disclaimer
2014 Research on Vocabulary Sizes and Codebook Universality
Wei-Xue Liu, Jian Hou, Hamid Reza Karimi
Abstr. Appl. Anal. 2014: 1-7 (2014). DOI: 10.1155/2014/697245

Abstract

Codebook is an effective image representation method. By clustering in local image descriptors, a codebook is shown to be a distinctive image feature and widely applied in object classification. In almost all existing works on codebooks, the building of the visual vocabulary follows a basic routine, that is, extracting local image descriptors and clustering with a user-designated number of clusters. The problem with this routine lies in that building a codebook for each single dataset is not efficient. In order to deal with this problem, we investigate the influence of vocabulary sizes on classification performance and vocabulary universality with the kNN classifier. Experimental results indicate that, under the condition that the vocabulary size is large enough, the vocabularies built from different datasets are exchangeable and universal.

Citation

Download Citation

Wei-Xue Liu. Jian Hou. Hamid Reza Karimi. "Research on Vocabulary Sizes and Codebook Universality." Abstr. Appl. Anal. 2014 1 - 7, 2014. https://doi.org/10.1155/2014/697245

Information

Published: 2014
First available in Project Euclid: 2 October 2014

zbMATH: 07022901
Digital Object Identifier: 10.1155/2014/697245

Rights: Copyright © 2014 Hindawi

JOURNAL ARTICLE
7 PAGES


SHARE
Vol.2014 • 2014
Back to Top