1.School of Automation and Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China
2.Center of National Railway Intelligent Transportation System Engineering and Technology,China Academy of Railway Sciences;Corporation Limited,Beijing 100081,China
In addressing the shortcomings of the existing methodologies for point cloud registration, including inadequate robustness, inaccurate feature representation and sensitivity to initial registration, particularly in the context of noisy point clouds, a novel point cloud registration network structure, RPM-AFAMNet, is proposed. This structure is constructed by integrating an adaptive feature aggregation module (AFAM) within the RPM-Net network. The AFAM is comprised of two modules: the batch attention mechanism transformer (BatchFormer) and the dynamic re-aggregated feature representation (DRFR) module. The purpose of the AFAM is twofold: firstly, to enhance the robustness of the network, and secondly, to optimise the representation of point cloud features. The BatchFormer is designed to mitigate the interference of noisy points by batch-weighted learning of point cloud features. The DRFR module, meanwhile, enhances the understanding of point cloud spatial relationships through a dynamic feature reorganisation strategy. This, in turn, improves the accuracy of point cloud alignment. Finally, the network is tested on the ModelNet40 dataset and significant improvements are achieved in six evaluation metrics, such as rotational and translational mean square error. Compared with RPM-Net, the following reductions are achieved: 32.05% for rotational mean square error, 75.86% for rotational mean absolute error, 33.33% for translational mean square error, 77.78% for translational mean absolute error, 33.03% for rotation angle error, and 36.36% for translation distance error. The experimental results demonstrate the efficacy of the proposed method in enhancing the registration performance of noise-containing point clouds.
YUJ Y, YANGY C, WANGY D,et al .An aerial photography slope point cloud filtering algorithm based on multi-plane segmentation and matrix transformations[J]. Journal of Hunan University (Natural Sciences),2024,51(11):12-22.(in Chinese)
WANGJ D, CAOJ, CHENZ G. Feature-preserving skeleton extraction algorithm for point clouds[J]. Journal of Graphics,2023,44(1):146-157.(in Chinese)
[5]
WEBERM, WILDD, KLEESIEKJ,et al .Deep learning-based point cloud registration for augmented reality-guided surgery[C]//2024 IEEE International Symposium on Biomedical Imaging (ISBI), May 27-30,2024,Athens,Greece. IEEE,2024:1-5.
QINX H, LIUS, CHANGD X,et al .LiDAR-GNSS/IMU automatic joint calibration method based on laser point cloud features[J].Journal of Hunan University (Natural Sciences),2023, 50(10): 175-184.(in Chinese)
LIUY Z, LIN, TAOZ Y .Point cloud classification and segmentation based on ring query and channel attention[J].Journal of Graphics,2022,43(4):616-623.(in Chinese)
YANGJ, GUOJ C .Recognition and segmentation of 3D point cloud by dynamic graph attention with adaptive generated convolutional kernel[J]. Journal of Hunan University (Natural Sciences), 2024, 51(12): 139-152.(in Chinese)
CHENGY J, QIUW G, LEIJ .Application of method for global extraction of tunnel centerlines based on 3D point clouds[J].Journal of Hunan University (Natural Sciences),2017,44(9):146-150.(in Chinese)
[16]
SHUY C, CHENGL J, XIAOB,et al .Focal-guided multi-consistency for unsupervised partial-to-partial point cloud registration[C]//2024 IEEE International Conference on Multimedia and Expo (ICME), July 15-19,2024,Niagara Falls,ON,Canada. IEEE,2024:1-6.
BAIC, CHENL, YANY. A RGB-D 3D point cloud registration method based on PVDAC descriptor[J]. Journal of Hunan University (Natural Sciences),2023,50(2):95-101.(in Chinese)
[19]
WUY, YAOQ L, FANX L,et al .PANet:a point-attention based multi-scale feature fusion network for point cloud registration[J].IEEE Transactions on Instrumentation and Measurement,2023,72: 1-13.
[20]
XUH, YEN J, LIUG H,et al. FINet:dual branches feature interaction for partial-to-partial point cloud registration[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2022, 36(3): 2848-2856.
[21]
HUANGX S, MEIG F, ZHANGJ .Feature-metric registration:a fast semi-supervised approach for robust point cloud registration without correspondences[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 13-19,2020.Seattle,WA,USA.IEEE,2020:11363-11371.
[22]
XIEY F, WANGB Y, LIS Q,et al .Iterative feedback network for unsupervised point cloud registration[J].IEEE Robotics and Automation Letters, 2024, 9(3): 2327-2334.
[23]
YEWZ J, LEEG H. RPM-net:robust point matching using learned features[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 13-19,2020.Seattle,WA,USA.IEEE,2020:11821-11830.
[24]
AOKIY, GOFORTHH, SRIVATSANR A,et al .PointNetLK:robust & efficient point cloud registration using PointNet[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 15-20,2019,Long Beach,CA,USA. IEEE,2020:7156-7165.
AOS, HUQ Y, WANGH Y,et al .BUFFER:balancing accuracy,efficiency,and generalizability in point cloud registration[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 17-24,2023,Vancouver,BC,Canada. IEEE,2023:1255-1264.
[27]
XUH, LIUS C, WANGG F,et al .OMNet:learning overlapping mask for partial-to-partial point cloud registration[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV),October 10-17,2021,Montreal,QC,Canada. IEEE, 2022:3112-3121.
[28]
FUK X, LUOJ Z, LUOX Y,et al .Robust point cloud registration framework based on deep graph matching[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(5): 6183-6191.
[29]
SHENQ B, ZHAOY G, YANGC,et al .3D pairwise point cloud registration network for power equipment based on multi-scale features[C]//2024 The 9th International Conference on Power and Renewable Energy (ICPRE), September 20-23,2024,Guangzhou, China. IEEE, 2024: 1452-1457.
[30]
MAL T, BIANW, XUEX P. Point clouds matching based on discrete optimal transport[J]. IEEE Transactions on Image Processing,2024,33:5650-5662.
[31]
WANGY, ZHOUP B, GENGG H,et al .MATR:multicompound adaptive transformer for point cloud registration[J]. IEEE Transactions on Industrial Informatics,2024,20(11):12903-12912.
[32]
HOUZ, YUB S, TAOD C .BatchFormer:learning to explore sample relationships for robust representation learning[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 18-24,2022,New Orleans,LA,USA. IEEE, 2022: 7246-7256.
[33]
LIH F, LIUJ Y, ZHANGY F,et al .A deep learning framework for infrared and visible image fusion without strict registration[J].International Journal of Computer Vision,2024,132(5):1625-1644.
[34]
WUY, ZHANGY, MAW P,et al .RORNet:partial-to-partial registration network with reliable overlapping representations[J].IEEE Transactions on Neural Networks and Learning Systems,2024,35(11): 15453-15466.