DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Yoon, Kuk-Jin | - |
dc.contributor.advisor | 윤국진 | - |
dc.contributor.author | Yoon, Sung-Hoon | - |
dc.date.accessioned | 2021-05-13T19:42:18Z | - |
dc.date.available | 2021-05-13T19:42:18Z | - |
dc.date.issued | 2020 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=947958&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/285223 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 기계공학과, 2020.2,[v,34 p. :] | - |
dc.description.abstract | In deep learning-based methods, many approaches, such as network architecture modification and input source diversification, have been applied to achieve performance improvement. However, existing methods have developed in their own aspect rather than taking advantage of other methods. Furthermore, to the best of my knowledge, no attempt has been made to merge the well performing networks. In this paper, I proposed a network that improves performance and reliability by using state-of-the-art networks as the baseline and fusing the results obtained by the baseline network. I validate the efficacy of the proposed fusion framework on the task of semantic segmentation | - |
dc.description.abstract | I compare the results from SOTA methods with that of the proposed SOTA-fusion framework. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Network calibration | - |
dc.subject | Semantic segmentation | - |
dc.subject | Deep learning | - |
dc.subject | Network fusion | - |
dc.subject | 네트워크 교정 | - |
dc.subject | 의미론적 영상분할 | - |
dc.subject | 딥러닝 | - |
dc.subject | 네트워크 융합 | - |
dc.title | Con-FusionNet: Multi-network fusion framework based on network calibration | - |
dc.title.alternative | 네트워크 교정에 기반한 다중 네트워크 융합 프레임 워크 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :기계공학과, | - |
dc.contributor.alternativeauthor | 윤성훈 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.