Unsupervised domain adaptation for object detection by controlling domain information도메인 정보 조절을 통한 물체 검출용 비지도 도메인 적응 기법
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Kim, Junmo | - |
| dc.contributor.advisor | 김준모 | - |
| dc.contributor.author | Yun, Woo Han | - |
| dc.date.accessioned | 2023-06-21T19:33:52Z | - |
| dc.date.available | 2023-06-21T19:33:52Z | - |
| dc.date.issued | 2022 | - |
| dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=996446&flag=dissertation | en_US |
| dc.identifier.uri | http://hdl.handle.net/10203/307953 | - |
| dc.description | 학위논문(박사) - 한국과학기술원 : 로봇공학학제전공, 2022.2,[v, 53 p. :] | - |
| dc.description.abstract | We discuss unsupervised domain adaptation for object detection by controlling domain information. To reduce the performance degradation by domain gaps between the source domain and target domain, we propose methods transferring domain information between two domains and expanding or shrinking data distribution in each domain. Through experiments, we verify that the proposed methods reduce the domain gap and improve the detection accuracy. | - |
| dc.language | eng | - |
| dc.publisher | 한국과학기술원 | - |
| dc.title | Unsupervised domain adaptation for object detection by controlling domain information | - |
| dc.title.alternative | 도메인 정보 조절을 통한 물체 검출용 비지도 도메인 적응 기법 | - |
| dc.type | Thesis(Ph.D) | - |
| dc.identifier.CNRN | 325007 | - |
| dc.description.department | 한국과학기술원 :로봇공학학제전공, | - |
| dc.contributor.alternativeauthor | 윤우한 | - |
- Appears in Collection
- RE-Theses_Ph.D.(박사논문)
- Files in This Item
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.