The limitations of point cloud registration networks in processing large-scale point clouds and capturing local detail features, resulted in insufficient registration accuracy for overlapping areas of point clouds, so a new point cloud registration network CR-RORNet was proposed.This network combined channel first convolutional attention and ResPointNet module, and overcame the shortcomings of existing methods in dealing with complex scenes and irregular point clouds, and improved the registration performance for point clouds with significant initial pose differences. Firstly, in the coarse registration stage, the ResPointNet module was designed to enhance the extraction and fusion of global and multi-level features of the point cloud model by introducing residual connection mechanism. Secondly, in the dynamic graph convolutional neural network, channel prior convolutional attention CPCA and cross stage gradient aggregation mechanism were combined. The CPCA mechanism utilized channel prior information to strengthen the network’s attention to important feature channels and regions. When dealing with overlapping parts of point clouds, it could effectively enhance the network model’s ability to capture local detail features of point clouds and suppress the influence of low confidence regions, significantly improving the registration effect; the cross stage gradient aggregation mechanism integrated gradient information from different depth levels of point cloud models, and ensured that the network could fully understand the structure and local details of the point cloud when dealing with small components or largescale scene point cloud models. And the learned features had good expressive power, so high-precision registration of point cloud data in complex scenes was realized. Experimental results show that CR-RORNet performs better than other point cloud registration methods on self collected datasets. Compared to baseline RORNet, CR-RORNet reduces error by 39.5% and error by 5.1%. Experiment results on the publicly available dataset ModelNet40 show that the network has good generalization performance.
ZHOUZ, FENGX, DIS, et al. A LiDAR mapping system for robot navigation in dynamic environments[J]. IEEE Transactions on Intelligent Vehicles, 2023:1-20.
[2]
ORTEGAS, SANTANAJ M, WENDELJ, et al. Generating 3D city models from open LiDAR point clouds: Advancing towards smart city applications[M]. Cham: Springer, 2020.
[3]
PARKJ, KIMC, KIMS, et al. PCSCNet: Fast 3D semantic segmentation of LiDAR point cloud for autonomous car using point convolution and sparse convolution network[J]. Expert Systems with Applications, 2023, 212:118815.
[4]
DUZ, YEH, CAOF. A novel local-global graph convolutional method for point cloud semantic segmentation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(4): 4798-4812.
[5]
LIS, LIUY, GALLJ. Rethinking 3-D LiDAR point cloud segmentation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021:1-12.
[6]
MISRAI, GIRDHARR, JOULINA. An end-to-end transformer model for 3d object detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 2886-2897.
[7]
CHENY, LIH, GAOR, et al. Boost 3-D object detection via point clouds segmentation and fused 3-D GIoU-L₁ loss[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022, 33(2): 762-773.
[8]
PHANA V, LE NGUYENM, NGUYENY L H, et al. DGCNN: A convolutional neural network over large-scale labeled graphs[J]. Neural Networks, 2018, 108: 533-543.
[9]
HUANGH, CHENZ, ZOUY, et al. Channel prior convolutional attention for medical image segmentation[DB/OL]. (2023-06-08)[2024-04-10].
[10]
WANGC Y, YEHI H, LIAOH Y M. YOLOv9: Learning what you want to learn using programmable gradient information[DB/OL]. (2024-02-21)[2024-04-10].
[11]
CHARLESR Q, HAOS, MOK, et al. PointNet: Deep learning on point sets for 3D classification and segmentation[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 77-85.
[12]
QIC R, YIL, SUH, et al. PointNet++: Deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017: 5105-5114.
[13]
ZHANGZ, DAIY, SUNJ. Deep learning based point cloud registration: An overview[J]. Virtual Reality & Intelligent Hardware, 2020, 2(3): 222-246.
[14]
AOKIY, GOFORTHH, SRIVATSANR A, et al. Pointnetlk: Robust & efficient point cloud registration using pointnet[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 7156-7165.
[15]
SARODEV, LIX, GOFORTHH, et al. PCRNet: Point cloud registration network using pointnet encoding[DB/OL]. (2019-08-21)[2024-04-10].
[16]
LUW, WANG, ZHOUY, et al. Deepvcp: An end-to-end deep neural network for point cloud registration[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 12-21.
[17]
PAISG D, RAMALINGAMS, GOVINDUV M, et al. 3DRegNet: A deep neural network for 3D point registration[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 7191-7201.
[18]
WUY, ZHANGY, MAW, et al. Rornet: Partial-to-partial registration network with reliable overlapping representations[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023: 1-14.
[19]
BAUERD, PATTENT, VINCZEM. Reagent: Point cloud registration using imitation and reinforcement learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 14581-14589.
[20]
CHENZ, CHENH, GONGL, et al. UTOPIC: Uncertainty‐aware overlap prediction network for partial point cloud registration[J]. Computer Graphics Forum, 2022, 41(7):87-98.
[21]
LIJ, ZHANGC, XUZ, et al. Iterative distance-aware similarity matrix convolution with mutual-supervised point elimination for efficient point cloud registration[C]//ECCV 2020: 16th European Conference, 2020: 378-394.
[22]
WANGY, SOLOMONJ M. Prnet: Self-supervised learning for partial-to-partial registration [DB/OL]. (2019-10-27)[2024-04-10].