DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 주재걸 | - |
dc.contributor.author | Choi, Dongmin | - |
dc.contributor.author | 최동민 | - |
dc.date.accessioned | 2024-07-30T19:30:38Z | - |
dc.date.available | 2024-07-30T19:30:38Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1096060&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/321355 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 김재철AI대학원, 2024.2,[iv, 29 p. :] | - |
dc.description.abstract | Accurately annotating multiple 3D objects in LiDAR scenes is laborious and challenging. While a few previous studies have attempted to leverage semi-automatic methods for cost-effective bounding box annotation, such methods have limitations in efficiently handling numerous multi-class objects. To effectively accelerate 3D annotation pipelines, we propose iDet3D, an efficient interactive 3D object detector. Supporting a user-friendly 2D interface, which can ease the cognitive burden of exploring 3D space to provide click interactions, iDet3D enables users to annotate the entire objects in each scene with minimal interactions. Taking the sparse nature of 3D point clouds into account, we design a negative click simulation to improve accuracy by reducing false-positive predictions. In addition, iDet3D incorporates two click propagation techniques to take full advantage of user interactions: (1) dense click guidance for keeping user-provided information throughout the network and (2) spatial click propagation for detecting other instances of the same class based on the user-specified objects. Through our extensive experiments, we present that our method can create precise annotations in a few clicks, which shows the practicality of iDet3D as an efficient annotation tool for 3D object detection. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 라이다▼a포인트 클라우드▼a3D 객체 탐지▼a사용자 상호작용 | - |
dc.subject | LiDAR▼aPoint cloud▼a3D object detection▼aUser interaction | - |
dc.title | Towards efficient interactive object detection for LiDAR point clouds | - |
dc.title.alternative | 라이다 포인트 클라우드를 위한 효율적인 상호 작용기반 물체 탐지 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :김재철AI대학원, | - |
dc.contributor.alternativeauthor | Choo, Jaegul | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.