DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 김민혁 | - |
dc.contributor.author | Choi, Dongyoung | - |
dc.contributor.author | 최동영 | - |
dc.date.accessioned | 2024-08-08T19:30:16Z | - |
dc.date.available | 2024-08-08T19:30:16Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1097311&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/321786 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전산학부, 2024.2,[iv, 32 p. :] | - |
dc.description.abstract | Omnidirectional cameras are extensively used in various applications to provide a wide field of vision. However, they face a challenge in synthesizing novel views due to the inevitable presence of dynamic objects, including the photographer, in their wide field of view. In this paper, we introduce a new approach called Omnidirectional Local Radiance Fields (OmniLocalRF) that can render static-only scene views, removing and inpainting dynamic objects simultaneously. Our approach combines the principles of local radiance fields with the bidirectional optimization of omnidirectional rays. Our input is an omnidirectional video, and we evaluate the mutual observations of the entire angle between the previous and current frames. To reduce ghosting artifacts of dynamic objects and inpaint occlusions, we devise a multi-resolution motion mask prediction module. Unlike existing methods that primarily separate dynamic components through the temporal domain, our method uses multi-resolution neural feature planes for precise segmentation, which is more suitable for long 360◦ videos. Our experiments validate that OmniLocalRF outperforms existing methods in both qualitative and quantitative metrics, especially in scenarios with complex real-world scenes. In particular, our approach eliminates the need for manual interaction, such as drawing motion masks by hand and additional pose estimation, making it a highly effective and efficient solution. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 전방향 카메라▼a가상 시점 영상 합성▼a뉴럴 광도 필드▼a양방향 최적화▼a이동체 마스크▼a카메라 행렬 추정 | - |
dc.subject | Omnidirectional camera▼aNovel view synthesis▼aNeural radiance fields▼aBidirectional optimization▼aMotion mask prediction▼aPose estimation | - |
dc.title | OmniLocalRF: Omnidirectional local radiance fields from dynamic videos | - |
dc.title.alternative | 동적 비디오에서의 가상 시점 영상 합성을 위한 전방향 로컬 광도 필드 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | Kim, Min Hyuk | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.