A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene

doi:10.3390/machines9100230

	A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene
	Liu, Huikai 1,2,3; Liu, Gaorui 1,3; Zhang, Yue 1,2,3; Lei, Linjian1,3,4 ; Xie, Hui 1,2,3; Li, Yan 1,2,3; Sun, Shengli1,3
	2021-10
发表期刊	MACHINES (IF:2.1[JCR-2023],2.2[5-Year])
EISSN	2075-1702
卷号	9 期号:10
DOI	10.3390/machines9100230
摘要	This paper addresses the problem of instance-level 6DoF pose estimation from a single RGBD image in an indoor scene. Many recent works have shown that a two-stage network, which first detects the keypoints and then regresses the keypoints for 6d pose estimation, achieves remarkable performance. However, the previous methods concern little about channel-wise attention and the keypoints are not selected by comprehensive use of RGBD information, which limits the performance of the network. To enhance RGB feature representation ability, a modular Split-Attention block that enables attention across feature-map groups is proposed. In addition, by combining the Oriented FAST and Rotated BRIEF (ORB) keypoints and the Farthest Point Sample (FPS) algorithm, a simple but effective keypoint selection method named ORB-FPS is presented to avoid the keypoints appear on the non-salient regions. The proposed algorithm is tested on the Linemod and the YCB-Video dataset, the experimental results demonstrate that our method outperforms the current approaches, achieves ADD(S) accuracy of 94.5% on the Linemod dataset and 91.4% on the YCB-Video dataset.
关键词	6DoF pose estimation split-channel attention ORB-FPS keypoint
URL	查看原文
收录类别	SCIE
语种	英语
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic ; Engineering, Mechanical
WOS记录号	WOS:000713310800001
出版者	MDPI
原始文献类型	Article
引用统计	正在获取...
文献类型	期刊论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/128596
专题	信息科学与技术学院_博士生信息科学与技术学院_特聘教授组_孙胜利组
通讯作者	Sun, Shengli
作者单位	1.Chinese Acad Sci, Shanghai Inst Tech Phys, Shanghai 200083, Peoples R China; 2.Univ Chinese Acad Sci, Sch Elect Elect & Commun Engn, Beijing 100049, Peoples R China; 3.Chinese Acad Sci, Key Lab Intelligent Infrared Percept, Shanghai 200083, Peoples R China; 4.ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai 201210, Peoples R China
推荐引用方式 GB/T 7714	Liu, Huikai,Liu, Gaorui,Zhang, Yue,et al. A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene[J]. MACHINES,2021,9(10).
APA	Liu, Huikai.,Liu, Gaorui.,Zhang, Yue.,Lei, Linjian.,Xie, Hui.,...&Sun, Shengli.(2021).A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene.MACHINES,9(10).
MLA	Liu, Huikai,et al."A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene".MACHINES 9.10(2021).