Detailed Information

Cited 2 time in webofscience Cited 3 time in scopus
Metadata Downloads

Improved Human-Object Interaction Detection Through On-the-Fly Stacked Generalization

Full metadata record
DC Field Value Language
dc.contributor.authorLee, Geonu-
dc.contributor.authorYun, Kimin-
dc.contributor.authorCho, Jungchan-
dc.date.available2021-03-15T01:40:33Z-
dc.date.created2021-03-15-
dc.date.issued2021-02-
dc.identifier.issn2169-3536-
dc.identifier.urihttps://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/80408-
dc.description.abstractHuman-object interaction (HOI) detection, which finds the relationships between humans and objects, is an important research area, but current HOI detection performance is unsatisfactory. One of the main problems is that CNN-based HOI detection algorithms fail to predict correct outputs for unseen test data based on a limited number of available training examples. Herein, we propose a novel framework for HOI detection called the on-the-fly stacked generalization deep neural network (OSGNet). OSGNet consists of three main components: (1) feature extraction modules, (2) HOI relationship detection networks, and (3) a meta-learner for combining the outputs of sub-models. Here, components (1) and (2) are considered to be sub-models. Any task-based feature extraction modules, such as classification or human pose estimation modules, can be used as sub-models. To achieve on-the-fly stacked generalization, the sub-models and meta-learner are trained simultaneously. The sub-models are trained to provide complementary information, and the meta-learner improves the generalization performance for unseen test data. Extensive experiments demonstrate that the proposed method achieves state-of-the-art accuracy, particularly in cases involving rare classes.-
dc.language영어-
dc.language.isoen-
dc.publisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC-
dc.relation.isPartOfIEEE ACCESS-
dc.titleImproved Human-Object Interaction Detection Through On-the-Fly Stacked Generalization-
dc.typeArticle-
dc.type.rimsART-
dc.description.journalClass1-
dc.identifier.wosid000626306900001-
dc.identifier.doi10.1109/ACCESS.2021.3061208-
dc.identifier.bibliographicCitationIEEE ACCESS, v.9, pp.34251 - 34263-
dc.description.isOpenAccessN-
dc.identifier.scopusid2-s2.0-85101760318-
dc.citation.endPage34263-
dc.citation.startPage34251-
dc.citation.titleIEEE ACCESS-
dc.citation.volume9-
dc.contributor.affiliatedAuthorLee, Geonu-
dc.contributor.affiliatedAuthorCho, Jungchan-
dc.type.docTypeArticle-
dc.subject.keywordAuthorFeature extraction-
dc.subject.keywordAuthorTask analysis-
dc.subject.keywordAuthorPose estimation-
dc.subject.keywordAuthorNeural networks-
dc.subject.keywordAuthorVisualization-
dc.subject.keywordAuthorTraining-
dc.subject.keywordAuthorStacking-
dc.subject.keywordAuthorDeep learning-
dc.subject.keywordAuthorhuman-object interaction-
dc.subject.keywordAuthorhuman pose estimation-
dc.subject.keywordAuthoraction recognition-
dc.relation.journalResearchAreaComputer Science-
dc.relation.journalResearchAreaEngineering-
dc.relation.journalResearchAreaTelecommunications-
dc.relation.journalWebOfScienceCategoryComputer Science, Information Systems-
dc.relation.journalWebOfScienceCategoryEngineering, Electrical & Electronic-
dc.relation.journalWebOfScienceCategoryTelecommunications-
dc.description.journalRegisteredClassscie-
dc.description.journalRegisteredClassscopus-
Files in This Item
There are no files associated with this item.
Appears in
Collections
IT융합대학 > 소프트웨어학과 > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Cho, Jung Chan photo

Cho, Jung Chan
College of IT Convergence (Department of Software)
Read more

Altmetrics

Total Views & Downloads

BROWSE