DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Munchurl | - |
dc.contributor.advisor | 김문철 | - |
dc.contributor.author | Moon, Jaeho | - |
dc.date.accessioned | 2023-06-26T19:33:36Z | - |
dc.date.available | 2023-06-26T19:33:36Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1033103&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/309825 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2023.2,[v, 38 p. :] | - |
dc.description.abstract | In this paper, we address the moving object problem in self-supervised learning of monocular depth estimation. A neural network learns to estimate depth maps based on the rigid scene assumption in self-supervised monocular depth estimation. However, moving objects that violate the rigid scene assumption induce incorrect depth estimation. We point out this problem comes from the pose network that only estimates the camera ego-motion between sequential frames. Thus, we suggest a Moving Object Disentangling Network, dubbed MODNet, that estimates the camera ego-motion as well as the object motion. To induce clear boundaries of the objects in motion estimation, we add an auxiliary branch to predict the binary foreground segmentation maps on the decoder of the motion estimation network. We show our MODNet resolves the moving object problem and improves the depth estimation performance with extensive experiments on the KITTI dataset. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Self-supervised learning▼amonocular depth estimation▼asemantic segmentation | - |
dc.subject | 자기 지도 학습▼a깊이 예측▼a의미론적 분할 | - |
dc.title | Self-supervised monocular depth estimation with learned-depth-prior-based disentanglement of moving objects from camera ego motion | - |
dc.title.alternative | 단시점 깊이 자기지도 학습을 위한 깊이 사전지식 기반 물체 움직임 분리 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 문재호 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.