File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Low-resolution image categorization via heterogeneous domain adaptation

TitleLow-resolution image categorization via heterogeneous domain adaptation
Authors
KeywordsHeterogeneous domain adaptation
Subspace learning
Low-resolution image categorization
Issue Date2019
Citation
Knowledge-Based Systems, 2019, v. 163, p. 656-665 How to Cite?
Abstract© 2018 Elsevier B.V. Most of existing image categorizations assume that the given datasets have a good resolution and quality. However, the assumption is often violated in real applications. In this paper, we study the low-resolution (LR) image categorization. By utilizing labeled high-resolution (HR) images as auxiliary information, we formulate the problem as a heterogeneous domain adaptation problem and propose a Discriminative Joint Distribution Adaptation (DJDA) model to solve it. The DJDA model projects both LR and HR images into an intermediate subspace with a well-designed objective function, where the distance between classes is expected to be enlarged and the distribution divergence to be reduced. As a result, the discriminative knowledge for HR images can be transferred effectively to LR images. Experimental results demonstrate the proposed DJDA method produces significantly superior categorization accuracies against state-of-the-art competitors.
Persistent Identifierhttp://hdl.handle.net/10722/277088
ISSN
2023 Impact Factor: 7.2
2023 SCImago Journal Rankings: 2.219
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorYao, Yuan-
dc.contributor.authorLi, Xutao-
dc.contributor.authorYe, Yunming-
dc.contributor.authorLiu, Feng-
dc.contributor.authorNg, Michael K.-
dc.contributor.authorHuang, Zhichao-
dc.contributor.authorZhang, Yu-
dc.date.accessioned2019-09-18T08:35:34Z-
dc.date.available2019-09-18T08:35:34Z-
dc.date.issued2019-
dc.identifier.citationKnowledge-Based Systems, 2019, v. 163, p. 656-665-
dc.identifier.issn0950-7051-
dc.identifier.urihttp://hdl.handle.net/10722/277088-
dc.description.abstract© 2018 Elsevier B.V. Most of existing image categorizations assume that the given datasets have a good resolution and quality. However, the assumption is often violated in real applications. In this paper, we study the low-resolution (LR) image categorization. By utilizing labeled high-resolution (HR) images as auxiliary information, we formulate the problem as a heterogeneous domain adaptation problem and propose a Discriminative Joint Distribution Adaptation (DJDA) model to solve it. The DJDA model projects both LR and HR images into an intermediate subspace with a well-designed objective function, where the distance between classes is expected to be enlarged and the distribution divergence to be reduced. As a result, the discriminative knowledge for HR images can be transferred effectively to LR images. Experimental results demonstrate the proposed DJDA method produces significantly superior categorization accuracies against state-of-the-art competitors.-
dc.languageeng-
dc.relation.ispartofKnowledge-Based Systems-
dc.subjectHeterogeneous domain adaptation-
dc.subjectSubspace learning-
dc.subjectLow-resolution image categorization-
dc.titleLow-resolution image categorization via heterogeneous domain adaptation-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.knosys.2018.09.027-
dc.identifier.scopuseid_2-s2.0-85054003702-
dc.identifier.volume163-
dc.identifier.spage656-
dc.identifier.epage665-
dc.identifier.isiWOS:000454468200052-
dc.identifier.issnl0950-7051-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats