DSpace university logo mark
Japanese | English 

NAOSITE : Nagasaki University's Academic Output SITE > 060 工学部・工学研究科 > 060 会議発表資料 >

Multi-label classification for image annotation via sparse similarity voting

ファイル 記述 サイズフォーマット
LNCS6469_344.pdf1.76 MBAdobe PDF本文ファイル

タイトル: Multi-label classification for image annotation via sparse similarity voting
著者: Sakai, Tomoya / Itoh, Hayato / Imiya, Atsushi
発行日: 2011年
出版者: Springer Verlag
引用: Lecture Notes in Computer Science, 6469(2), pp.344-353; 2011
抄録: We present a supervised multi-label classification method for automatic image annotation. Our method estimates the annotation labels for a test image by accumulating similarities between the test image and labeled training images. The similarities are measured on the basis of sparse representation of the test image by the training images, which avoids similarity votes for irrelevant classes. Besides, our sparse representation-based multi-label classification can estimate a suitable combination of labels even if the combination is unlearned. Experimental results using the PASCAL dataset suggest effectiveness for image annotation compared to the existing SVM-based multi-labeling methods. Nonlinear mapping of the image representation using the kernel trick is also shown to enhance the annotation performance.
記述: International Workshops on Computer Vision, ACCV 2010; Queenstown; 8 November 2010 through 9 November 2010
URI: http://hdl.handle.net/10069/27087
ISSN: 03029743
権利: © 2011 Springer-Verlag Berlin Heidelberg. / The original publication is available at www.springerlink.com
資料タイプ: Conference Paper
原稿種類: author
出現コレクション:060 会議発表資料

引用URI : http://hdl.handle.net/10069/27087



Valid XHTML 1.0! Copyright © 2006-2015 長崎大学附属図書館 - お問い合わせ Powerd by DSpace