論文

査読有り
2010年

SHRINKING LARGE VISUAL VOCABULARIES USING MULTI-LABEL AGGLOMERATIVE INFORMATION BOTTLENECK

2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING
  • Wojciech Wojcikiewicz
  • ,
  • Alexander Binder
  • ,
  • Motoaki Kawanabe

開始ページ
3849
終了ページ
3852
記述言語
英語
掲載種別
研究論文(国際会議プロシーディングス)
DOI
10.1109/ICIP.2010.5653575
出版者・発行元
IEEE

The quality of visual vocabularies is crucial for the performance of bag-of-words image classification methods. Several approaches have been developed for codebook construction, the most popular method is to cluster a set of image features (e.g. SIFT) by k-means. In this paper, we propose a two-step procedure which incorporates label information into the clustering process by efficiently generating a large and informative vocabulary using class-wise k-means and reducing its size by agglomerative information bottleneck (AIB). We introduce an extension of the AIB procedure for multi-label problems and show that this two-step approach improves the classification results while reducing computation time compared to the vanilla k-means. We analyse the reasons for the performance gain on the PASCAL VOC 2007 data set.

Web of Science ® 被引用回数 : 1

リンク情報
DOI
https://doi.org/10.1109/ICIP.2010.5653575
Web of Science
https://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=JSTA_CEL&SrcApp=J_Gate_JST&DestLinkType=FullRecord&KeyUT=WOS:000287728003227&DestApp=WOS_CPL
URL
http://dblp.uni-trier.de/db/conf/icip/icip2010.html#conf/icip/WojcikiewiczBK10
ID情報
  • DOI : 10.1109/ICIP.2010.5653575
  • ISSN : 1522-4880
  • DBLP ID : conf/icip/WojcikiewiczBK10
  • Web of Science ID : WOS:000287728003227

エクスポート
BibTeX RIS