DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Dae-Shik | - |
dc.contributor.advisor | 김대식 | - |
dc.contributor.author | Kim, Bomi | - |
dc.date.accessioned | 2023-06-26T19:33:33Z | - |
dc.date.available | 2023-06-26T19:33:33Z | - |
dc.date.issued | 2022 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=997178&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/309815 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2022.2,[iii, 19 p. :] | - |
dc.description.abstract | Color conveys important information about the visible world. However, under low-light conditions, both pixel intensity, as well as true color distribution, can be significantly shifted. Moreover, most of such distortions are non-recoverable due to inverse problems. In the present study, we utilized recent advancements in learning-based methods for low-light image enhancement. However, while most ``deep learning" methods aim to restore high-level and object-oriented visual information, we hypothesized that learning-based methods can also be used for restoring color-based information. To address this question, we propose a novel color representation learning method for low-light image enhancement. More specifically, we used a channel-aware residual network and a differentiable intensity histogram to capture color features. Experimental results using synthetic and natural datasets suggest that the proposed learning scheme achieves state-of-the-art performance. We conclude from our study that inter-channel dependency and color distribution matching are crucial factors for learning color representations under low-light conditions. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.title | Learning color representations for low-light image enhancement | - |
dc.title.alternative | 저조도 이미지 복원을 위한 색상 정보 학습 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 김보미 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.