Active reranking for web image search

Image search reranking methods usually fail to capture the user's intention when the query term is ambiguous. Therefore, reranking with user interactions, or active reranking, is highly demanded to effectively improve the search performance. The essential problem in active reranking is how to t...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on image processing : a publication of the IEEE Signal Processing Society. - 1992. - 19(2010), 3 vom: 19. März, Seite 805-20
1. Verfasser: Tian, Xinmei (VerfasserIn)
Weitere Verfasser: Tao, Dacheng, Hua, Xian-Sheng, Wu, Xiuqing
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2010
Zugriff auf das übergeordnete Werk:IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
Schlagworte:Journal Article Research Support, Non-U.S. Gov't
LEADER 01000naa a22002652 4500
001 NLM192590367
003 DE-627
005 20231223193302.0
007 cr uuu---uuuuu
008 231223s2010 xx |||||o 00| ||eng c
024 7 |a 10.1109/TIP.2009.2035866  |2 doi 
028 5 2 |a pubmed24n0642.xml 
035 |a (DE-627)NLM192590367 
035 |a (NLM)19887316 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Tian, Xinmei  |e verfasserin  |4 aut 
245 1 0 |a Active reranking for web image search 
264 1 |c 2010 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Completed 11.05.2010 
500 |a Date Revised 22.02.2010 
500 |a published: Print-Electronic 
500 |a Citation Status PubMed-not-MEDLINE 
520 |a Image search reranking methods usually fail to capture the user's intention when the query term is ambiguous. Therefore, reranking with user interactions, or active reranking, is highly demanded to effectively improve the search performance. The essential problem in active reranking is how to target the user's intention. To complete this goal, this paper presents a structural information based sample selection strategy to reduce the user's labeling efforts. Furthermore, to localize the user's intention in the visual feature space, a novel local-global discriminative dimension reduction algorithm is proposed. In this algorithm, a submanifold is learned by transferring the local geometry and the discriminative information from the labelled images to the whole (global) image database. Experiments on both synthetic datasets and a real Web image search dataset demonstrate the effectiveness of the proposed active reranking scheme, including both the structural information based active sample selection strategy and the local-global discriminative dimension reduction algorithm 
650 4 |a Journal Article 
650 4 |a Research Support, Non-U.S. Gov't 
700 1 |a Tao, Dacheng  |e verfasserin  |4 aut 
700 1 |a Hua, Xian-Sheng  |e verfasserin  |4 aut 
700 1 |a Wu, Xiuqing  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on image processing : a publication of the IEEE Signal Processing Society  |d 1992  |g 19(2010), 3 vom: 19. März, Seite 805-20  |w (DE-627)NLM09821456X  |x 1941-0042  |7 nnns 
773 1 8 |g volume:19  |g year:2010  |g number:3  |g day:19  |g month:03  |g pages:805-20 
856 4 0 |u http://dx.doi.org/10.1109/TIP.2009.2035866  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 19  |j 2010  |e 3  |b 19  |c 03  |h 805-20