Presentation 2011-02-18
Negative Selection for High Dimensional Feature in Video Retrieval with Query-by-Example Method
Yuta MATSUOKA, Kimiaki SHIRAHAMA, Kuniaki UEHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The query-by-example approach can be considered as partially supervised learning, where a retrieval model for a query is extrated by using positive examples provided by a user and unlabeled examples. For accurate retrieval, it is crucial how to select negative examples from unlabeled examples. We select negative examples which are highly similar to positive examples, because they are useful for denning the boundary between relevant and irrelevant shots. Such negative examples are selected by iteratively building SVMs, which are used to examine whether unlabeled examples are dissimilar to positive examples. The proposed method is tested on TRECVID 2009 video data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Query by example / partially supervised learning / negative selection / TRECVID
Paper # PRMU2010-227
Date of Issue

Conference Information
Committee PRMU
Conference Date 2011/2/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Negative Selection for High Dimensional Feature in Video Retrieval with Query-by-Example Method
Sub Title (in English)
Keyword(1) Query by example
Keyword(2) partially supervised learning
Keyword(3) negative selection
Keyword(4) TRECVID
1st Author's Name Yuta MATSUOKA
1st Author's Affiliation Graduate School of Engineering, Kobe University()
2nd Author's Name Kimiaki SHIRAHAMA
2nd Author's Affiliation Graduate School of Economics, Kobe University
3rd Author's Name Kuniaki UEHARA
3rd Author's Affiliation Graduate School of Informatics, Kobe University
Date 2011-02-18
Paper # PRMU2010-227
Volume (vol) vol.110
Number (no) 414
Page pp.pp.-
#Pages 6
Date of Issue