SelfReg: Self-supervised Contrastive Regularization for Domain Generalization
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, D. | - |
dc.contributor.author | Yoo, Y. | - |
dc.contributor.author | Park, S. | - |
dc.contributor.author | Kim, J. | - |
dc.contributor.author | Lee, J. | - |
dc.date.accessioned | 2022-12-12T05:41:23Z | - |
dc.date.available | 2022-12-12T05:41:23Z | - |
dc.date.created | 2022-12-08 | - |
dc.date.issued | 2021 | - |
dc.identifier.issn | 1550-5499 | - |
dc.identifier.uri | https://scholar.korea.ac.kr/handle/2021.sw.korea/147155 | - |
dc.description.abstract | In general, an experimental environment for deep learning assumes that the training and the test dataset are sampled from the same distribution. However, in real-world situations, a difference in the distribution between two datasets, i.e. domain shift, may occur, which becomes a major factor impeding the generalization performance of the model. The research field to solve this problem is called domain generalization, and it alleviates the domain shift problem by extracting domain-invariant features explicitly or implicitly. In recent studies, contrastive learning-based domain generalization approaches have been proposed and achieved high performance. These approaches require sampling of the negative data pair. However, the performance of contrastive learning fundamentally depends on quality and quantity of negative data pairs. To address this issue, we propose a new regularization method for domain generalization based on contrastive learning, called self-supervised contrastive regularization (SelfReg). The proposed approach use only positive data pairs, thus it resolves various problems caused by negative pair sampling. Moreover, we propose a class-specific domain perturbation layer (CDPL), which makes it possible to effectively apply mixup augmentation even when only positive data pairs are used. The experimental results show that the techniques incorporated by SelfReg contributed to the performance in a compatible manner. In the recent benchmark, DomainBed, the proposed method shows comparable performance to the conventional state-of-the-art alternatives. © 2021 IEEE | - |
dc.language | English | - |
dc.language.iso | en | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | SelfReg: Self-supervised Contrastive Regularization for Domain Generalization | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Kim, J. | - |
dc.identifier.doi | 10.1109/ICCV48922.2021.00948 | - |
dc.identifier.scopusid | 2-s2.0-85127758384 | - |
dc.identifier.bibliographicCitation | Proceedings of the IEEE International Conference on Computer Vision, pp.9599 - 9608 | - |
dc.relation.isPartOf | Proceedings of the IEEE International Conference on Computer Vision | - |
dc.citation.title | Proceedings of the IEEE International Conference on Computer Vision | - |
dc.citation.startPage | 9599 | - |
dc.citation.endPage | 9608 | - |
dc.type.rims | ART | - |
dc.type.docType | Conference Paper | - |
dc.description.journalClass | 1 | - |
dc.description.journalRegisteredClass | scopus | - |
dc.subject.keywordAuthor | Domain Generalization | - |
dc.subject.keywordAuthor | Image Recognition | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
(02841) 서울특별시 성북구 안암로 14502-3290-1114
COPYRIGHT © 2021 Korea University. All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.