Unsupervised non-rigid point cloud registration based on point-wise displacement learning

被引:0
作者
Yiqi Wu
Fang Han
Dejun Zhang
Tiantian Zhang
Yilin Chen
机构
[1] China University of Geosciences,School of Computer Science
[2] Wuhan Institute of Technology,Hubei Key Laboratory of Intelligent Robot
来源
Multimedia Tools and Applications | 2024年 / 83卷
关键词
Point cloud; Non-rigid registration; Point displacement; Self attention; Deep learning;
D O I
暂无
中图分类号
学科分类号
摘要
Registration of deformable objects is a fundamental prerequisite for many modern virtual reality and computer vision applications. However, due to the difficulties of acquiring labeled datasets and the inherent irregular deformation, non-rigid registration for 3D scanner-captured data remains challenging. This paper proposes an unsupervised non-rigid 3D point cloud registration network based on the self-attention mechanism. Specifically, considering the registration as the result of point drifts between the source and target shapes, a Transformer-based encoder-decoder module is utilized to estimate the point displacements. Additionally, a symmetric registration procedure is adopted with regularization loss to manage the regular deformation of points, ultimately producing reasonable registration results for real-world deformable objects. Experiments are conducted on public and synthesized datasets which simulate diversiform non-rigid 2D or 3D deformations. Numerical and qualitative experimental results demonstrate that the proposed network achieves outstanding performance and is robust in scenes with multiple interferences.
引用
收藏
页码:24589 / 24607
页数:18
相关论文
共 50 条
[41]   A Comparative Study of Downsampling Techniques for Non-rigid Point Set Registration Using Color [J].
Saval-Calvo, Marcelo ;
Orts-Escolano, Sergio ;
Azorin-Lopez, Jorge ;
Garcia Rodriguez, Jose ;
Fuster-Guillo, Andres ;
Morell-Gimenez, Vicente ;
Cazorla, Miguel .
BIOINSPIRED COMPUTATION IN ARTIFICIAL SYSTEMS, PT II, 2015, 9108 :281-290
[42]   Non-rigid registration of 3D point clouds under isometric deformation [J].
Ge, Xuming .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2016, 121 :192-202
[43]   Deep learning based point cloud registration: an overview [J].
Zhang Z. ;
Dai Y. ;
Sun J. .
Dai, Yuchao (daiyuchao@nwpu.edu.cn), 1600, KeAi Communications Co. (02) :222-246
[44]   Point set non-rigid registration using t-distribution mixture model [J].
Zhang, T. (zhangt@ciomp.ac.cn), 2013, Chinese Academy of Sciences (21) :2405-2420
[45]   Medical image non-rigid registration method guided by a matched feature point pair [J].
Wang, Zhen-Song ;
Liu, Xiao-Yun ;
Chen, Wu-Fan .
Dianzi Keji Daxue Xuebao/Journal of the University of Electronic Science and Technology of China, 2012, 41 (03) :471-476
[46]   Deep learning-based point cloud registration: a comprehensive investigation [J].
Cheng, Xiaolong ;
Liu, Xinyu ;
Li, Jintao ;
Zhou, Wei .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2024, 45 (10) :3412-3442
[47]   Deep Graph Matching Based Dense Correspondence Learning Between Non-Rigid Point Clouds [J].
Luo, Jiazheng ;
Yuan, Mingzhi ;
Fu, Kexue ;
Wang, Manning ;
Zhang, Chenxi .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2022, 7 (03) :5842-5849
[48]   Automated method for measuring body size parameters of live pigs based on non-rigid registration of point clouds [J].
Gao, Zicheng ;
Lei, Jie ;
Wu, Jianhuan ;
Zhang, Jialong ;
Ruchay, Alexey ;
Pezzuolo, Andrea ;
Guo, Hao .
PROCEEDINGS OF 2023 IEEE INTERNATIONAL WORKSHOP ON METROLOGY FOR AGRICULTURE AND FORESTRY, METROAGRIFOR, 2023, :472-477
[49]   Volume reconstruction based on non-rigid registration [J].
Bao, Xudong ;
Xu, Danhua ;
Toumoulin, Christine ;
Luo, Limin .
2007 ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-16, 2007, :6536-+
[50]   Equivariance-Based Markov Decision Process for Unsupervised Point Cloud Registration [J].
Wu, Yue ;
Lei, Jiayi ;
Yuan, Yongzhe ;
Fan, Xiaolong ;
Gong, Maoguo ;
Ma, Wenping ;
Miao, Qiguang ;
Zhang, Mingyang .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2025, 35 (05) :4648-4660