File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Improving web image search by bag-based reranking

TitleImproving web image search by bag-based reranking
Authors
KeywordsBag-based image reranking
generalized multi-instance (GMI) learning
text-based image retrieval (TBIR)
Issue Date2011
Citation
IEEE Transactions on Image Processing, 2011, v. 20, n. 11, p. 3280-3290 How to Cite?
AbstractGiven a textual query in traditional text-based image retrieval (TBIR), relevant images are to be reranked using visual features after the initial text-based search. In this paper, we propose a new bag-based reranking framework for large-scale TBIR. Specifically, we first cluster relevant images using both textual and visual features. By treating each cluster as a bag and the images in the bag as instances, we formulate this problem as a multi-instance (MI) learning problem. MI learning methods such as mi-SVM can be readily incorporated into our bag-based reranking framework. Observing that at least a certain portion of a positive bag is of positive instances while a negative bag might also contain positive instances, we further use a more suitable generalized MI (GMI) setting for this application. To address the ambiguities on the instance labels in the positive and negative bags under this GMI setting, we develop a new method referred to as GMI-SVM to enhance retrieval performance by propagating the labels from the bag level to the instance level. To acquire bag annotations for (G)MI learning, we propose a bag ranking method to rank all the bags according to the defined bag ranking score. The top ranked bags are used as pseudopositive training bags, while pseudonegative training bags can be obtained by randomly sampling a few irrelevant images that are not associated with the textual query. Comprehensive experiments on the challenging real-world data set NUS-WIDE demonstrate our framework with automatic bag annotation can achieve the best performances compared with existing image reranking methods. Our experiments also demonstrate that GMI-SVM can achieve better performances when using the manually labeled training bags obtained from relevance feedback. © 2011 IEEE.
Persistent Identifierhttp://hdl.handle.net/10722/321448
ISSN
2023 Impact Factor: 10.8
2023 SCImago Journal Rankings: 3.556
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorDuan, Lixin-
dc.contributor.authorLi, Wen-
dc.contributor.authorTsang, Ivor Wai Hung-
dc.contributor.authorXu, Dong-
dc.date.accessioned2022-11-03T02:19:00Z-
dc.date.available2022-11-03T02:19:00Z-
dc.date.issued2011-
dc.identifier.citationIEEE Transactions on Image Processing, 2011, v. 20, n. 11, p. 3280-3290-
dc.identifier.issn1057-7149-
dc.identifier.urihttp://hdl.handle.net/10722/321448-
dc.description.abstractGiven a textual query in traditional text-based image retrieval (TBIR), relevant images are to be reranked using visual features after the initial text-based search. In this paper, we propose a new bag-based reranking framework for large-scale TBIR. Specifically, we first cluster relevant images using both textual and visual features. By treating each cluster as a bag and the images in the bag as instances, we formulate this problem as a multi-instance (MI) learning problem. MI learning methods such as mi-SVM can be readily incorporated into our bag-based reranking framework. Observing that at least a certain portion of a positive bag is of positive instances while a negative bag might also contain positive instances, we further use a more suitable generalized MI (GMI) setting for this application. To address the ambiguities on the instance labels in the positive and negative bags under this GMI setting, we develop a new method referred to as GMI-SVM to enhance retrieval performance by propagating the labels from the bag level to the instance level. To acquire bag annotations for (G)MI learning, we propose a bag ranking method to rank all the bags according to the defined bag ranking score. The top ranked bags are used as pseudopositive training bags, while pseudonegative training bags can be obtained by randomly sampling a few irrelevant images that are not associated with the textual query. Comprehensive experiments on the challenging real-world data set NUS-WIDE demonstrate our framework with automatic bag annotation can achieve the best performances compared with existing image reranking methods. Our experiments also demonstrate that GMI-SVM can achieve better performances when using the manually labeled training bags obtained from relevance feedback. © 2011 IEEE.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Image Processing-
dc.subjectBag-based image reranking-
dc.subjectgeneralized multi-instance (GMI) learning-
dc.subjecttext-based image retrieval (TBIR)-
dc.titleImproving web image search by bag-based reranking-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TIP.2011.2159227-
dc.identifier.pmid21659029-
dc.identifier.scopuseid_2-s2.0-80054826786-
dc.identifier.volume20-
dc.identifier.issue11-
dc.identifier.spage3280-
dc.identifier.epage3290-
dc.identifier.isiWOS:000296016200023-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats