DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Woo, Woontack | - |
dc.contributor.advisor | 우운택 | - |
dc.contributor.author | Kim, Jeonghyun | - |
dc.date.accessioned | 2023-06-22T19:31:51Z | - |
dc.date.available | 2023-06-22T19:31:51Z | - |
dc.date.issued | 2022 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1021032&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/308298 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 문화기술대학원, 2022.8,[v, 38 p. :] | - |
dc.description.abstract | We propose Seg&Struct, a supervised learning framework leveraging the interplay between part segmentation and structure inference and demonstrating their synergy in an integrated framework. Both 3D segmentation and structure inference have been extensively studied in the recent deep learning literature, while the supervisions used for each task have not been fully exploited to assist the other task. Namely, structure inference has been typically conducted with an autoencoder that does not leverage the point-to-part associations. Also, 3D segmentation has been mostly performed without structural priors that tell the plausibility of the structure created from the output segments. We present how these two tasks can be best combined while fully utilizing supervision to improve performance. Our framework first decomposes a raw input shape into segments using an off-the-shelf segmentation network, whose outputs are then mapped to nodes in a part hierarchy. Following this, a hierarchical message passing based on the structural priors regresses the parameters of part bounding boxes and classified the part relationships. Once the structure is predicted, the segmentation is rectified by examining the confusion of part boundaries using the structure-based part features. The refined segmentation can again be used to improve the predicted structure. Our experimental results based on the StructureNet dataset demonstrate that the interplay between the two tasks results in significant improvements in both tasks: 30.94% in structure inference and 0.6% in segmentation. We also showcase a structure-based shape retrieval that enables finding semantically similar shapes. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Shape Analysis▼aArtificial Intelligence▼aPart Segmentation▼aStructure Decomposition▼a3D Computer Vision▼a3D Computer Graphics | - |
dc.subject | 형상 분석▼a인공지능▼a형상 부분 분할▼a형상 구조 분해▼a3차원 컴퓨터 비전▼a3차원 컴퓨터 그래픽스 | - |
dc.title | (The) interplay between part segmentation and structure inference for 3D shape parsing | - |
dc.title.alternative | 3차원 형상 분해를 위한 형상 부분 분할과 구조 추론의 상호작용 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :문화기술대학원, | - |
dc.contributor.alternativeauthor | 김정현 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.