DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kang, Dongmin | ko |
dc.contributor.author | Nam, Yeongwoo | ko |
dc.contributor.author | Kyung, Daeun | ko |
dc.contributor.author | Choi, Jonghyun | ko |
dc.date.accessioned | 2022-06-21T10:02:52Z | - |
dc.date.available | 2022-06-21T10:02:52Z | - |
dc.date.created | 2022-06-21 | - |
dc.date.created | 2022-06-21 | - |
dc.date.issued | 2022 | - |
dc.identifier.citation | IEEE ACCESS, v.10, pp.56901 - 56913 | - |
dc.identifier.issn | 2169-3536 | - |
dc.identifier.uri | http://hdl.handle.net/10203/297025 | - |
dc.description.abstract | Input-level domain adaptation reduces the burden of a neural encoder without supervision by reducing the domain gap at the input level. Input-level domain adaptation is widely employed in 2D visual domain, e.g., images and videos, but is not utilized for 3D point clouds. We propose the use of input-level domain adaptation for 3D point clouds, namely, point-level domain adaptation. Specifically, we propose to learn a transformation of 3D point clouds by searching the best combination of operations on point clouds that transfer data from the source domain to the target domain while maintaining the classification label without supervision of the target label. We decompose the learning objective into two terms, resembling domain shift and preserving label information. On the PointDA-10 benchmark dataset, our method outperforms state-of-the-art, unsupervised, point cloud domain adaptation methods by large margins (up to + 3.97 % in average). | - |
dc.language | English | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.title | Unsupervised Domain Adaptation for 3D Point Clouds by Searched Transformations | - |
dc.type | Article | - |
dc.identifier.wosid | 000805793700001 | - |
dc.identifier.scopusid | 2-s2.0-85130444676 | - |
dc.type.rims | ART | - |
dc.citation.volume | 10 | - |
dc.citation.beginningpage | 56901 | - |
dc.citation.endingpage | 56913 | - |
dc.citation.publicationname | IEEE ACCESS | - |
dc.identifier.doi | 10.1109/ACCESS.2022.3176719 | - |
dc.contributor.localauthor | Kyung, Daeun | - |
dc.contributor.nonIdAuthor | Kang, Dongmin | - |
dc.contributor.nonIdAuthor | Nam, Yeongwoo | - |
dc.contributor.nonIdAuthor | Choi, Jonghyun | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Point cloud compression | - |
dc.subject.keywordAuthor | Three-dimensional displays | - |
dc.subject.keywordAuthor | Adaptation models | - |
dc.subject.keywordAuthor | Training | - |
dc.subject.keywordAuthor | Feature extraction | - |
dc.subject.keywordAuthor | Data models | - |
dc.subject.keywordAuthor | Search problems | - |
dc.subject.keywordAuthor | Data transformation | - |
dc.subject.keywordAuthor | domain adaptation | - |
dc.subject.keywordAuthor | point cloud recognition | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.