计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (11A): 250100019-7.doi: 10.11896/jsjkx.250100019

• 计算机图形学&多媒体 • 上一篇    下一篇

融合自适应优化与多维聚焦的点云配准网络

岳倩雯1, 王东强2, 张强1   

  1. 1 重庆理工大学两江人工智能学院 重庆 401135
    2 中交第一公路勘察设计研究院有限公司 西安 710075
  • 出版日期:2025-11-15 发布日期:2025-11-10
  • 通讯作者: 王东强(48001416@qq.com)
  • 作者简介:3045397322@qq.com
  • 基金资助:
    中交集团数字化重大专项(KYHT2023-113)

Point Cloud Registration Network Integrating Adaptive Optimization and Multi-dimensional Focusing

YUE Qianwen1, WANG Dongqiang2, ZHANG Qiang1   

  1. 1 Department of Artificial Intelligence,Chongqing University of Technology,Chongqing 401135,China
    2 CCCC FIRST HIGHWAY CONSULTANTS CO.LTD,Xi’an 710075,China
  • Online:2025-11-15 Published:2025-11-10
  • Supported by:
    Research and Demonstration of Digital Application Technology for Key City Business(KYHT2023-113).

PDF (PC)

10

摘要： 在点云配准领域,面对低重叠率的点云时,如何有效捕捉细节特征并提高配准精度是两大核心挑战。尽管传统特征提取方法已取得一定成效,但其对点云几何信息的挖掘尚显不足,导致所提取特征的区分度有限。当前的技术主要依赖于位置编码和几何嵌入策略,虽在一定程度上增强了模型的几何理解能力,但在面对高离群值的场景时,配准精度仍有提升空间。为了解决这些问题,提出了一种融合自适应优化与多维聚焦的点云配准网络ROPNet。通过引入多维聚焦、自适应调制核以及动态优化选择器模块来捕捉全局特征和局部细节,识别点云的空间位置和对应关系,理解点云数据的内在结构,增强内点的识别能力,提升了配准精度。实验结果表明,ROPNet在多个数据集上均展现出优越性能。具体来说,在3DMatch数据集中,将配准召回率提升至92.4%,内点比率提高到71.3%。而在KITTI数据集上,不仅实现99.8%的高配准精度,同时还将相对旋转误差降低至0.24°,相对平移误差降低至6.6 cm。

关键词: 点云配准, 多维聚焦, 深度学习, 特征提取, 自适应优化

Abstract: In the field of point cloud registration,effectively capturing detailed features and enhancing registration accuracy,particularly when handling point clouds with low overlap rates,are two primary challenges.Traditional feature extraction methods have achieved some success but remain insufficient in mining geometric information,leading to limited feature discrimination.Current techniques primarily rely on position encoding and geometric embedding strategies,which enhance geometric understanding but still struggle with accuracy in high-outlier scenes.To address these issues,this paper introduces ROPNet,a novel registration network that integrates adaptive optimization and multi-dimensional focusing.ROPNet’s design includes multi-dimensional focusing,adaptive modulation kernels,and a dynamic optimization selector.These components enable the network to capture both global features and local details,accurately identify spatial positions and correspondences,and better understand the intrinsic structure of point cloud data.Additionally,ROPNet’s robust design significantly improves the identification of inliers,thereby significantly improving registration accuracy.Experimental results demonstrate ROPNet’s superior performance.On the 3DMatch dataset,ROPNet achieves a 92.4% registration recall rate and a 71.3% inlier ratio.On the KITTI dataset,it attains 99.8% registration accuracy,with relative rotation and translation errors reduces to 0.24 degrees and 6.6 cm,respectively.

Key words: Point cloud registration, Multi-dimensional focusing, Deep learning, Feature extraction, Adaptive optimization

中图分类号: 

  • TP391.41
[1]SCHNABEL R,WAHL R,KLEIN R.Efficient ransac for point-cloud shape detection[C]//Computer Graphics Forum.2007:214-226.
[2]AIGER D,MITRA N J,COHEN-OR D.4-points congruent sets for robust pairwise surface registration[C]//ACM SIGGRAPH 2008.2008:1-10.
[3]MELLADO N,AIGER D,MITRA N J.Super 4pcs fast globalpointcloud registration via smart indexing[C]//Computer Graphics Forum.2014:205-215.
[4]SCHENKER P S.SENSOR FUSION IV:Control paradigms and data structures[C]//Proceedings of the Meeting,Boston,Ma(Number SPIE-1611).Society of Photo-Optical Instrumentation Engineers(SPIE Proceedings.Vol.1611),1992.
[5]MAGNUSSON M.The three dimensional normal-distributionstransform:an efficient representation for registration,surface analysis,and loop detection[D].Orebro universitet,2009.
[6]SHARP G C,LEE S W,WEHED K.Icp registration using invariant features[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(1):90-102.
[7]BOUAZIZ S,AGLIASACCHI A,PAULY M.Sparse iterativeclosest point[C]//Computer Graphics Forum.2013:113-123.
[8]YANG J,LI H,CAMPBELL D,et al.Go-icp:A globally optimal solution to 3d icp point-set registration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(11):2241-2254.
[9]RUSU R B,BLODOW N,BEETZ M.Fast point feature histograms(fpfh) for 3d registration[C]//2009 IEEE International Conference on Robotics and Automation.IEEE,2009:3212-3217.
[10]SALTI S,TOMBARI F,DI STEFANO L.Shot:Unique signatures of histograms for surface and texture description[J].Computer Vision and Image Understanding,2014,125:251-264.
[11]CHOY C,PARK J,KOLTUN V.Fully convolutional geometric features[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:8958-8966.
[12]BAI X,LUO Z,ZHOU L,et al.D3feat:Joint learning of dense detection and description of 3d local features[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:6359-6367.
[13]AOKI Y,GOFORTH H,SRIVATSAN R A,et al.Pointnetlk:Robust & efficient point cloud registration using pointnet[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:7163-7172.
[14]BAKER S,MATTHEWS I.Lucas-kanade 20 years on:A unifying framework[J].International Journal of Computer Vision,2004,56:221-255.
[15]QI C R,SU H,MO K,et al.Pointnet:Deep learning on point sets for 3d classification and segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:652-660.
[16]WANG Y,SOLOMON J M.Deep closest point:Learning representations for point cloud registration[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:3523-3532.
[17]HUANG S,GOJCIC Z,USVYATSOV M,et al.Predator:Re-gistration of 3d point clouds with low overlap[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:4267-4276.
[18]HUANG X,QU W,ZUO Y,et al.Imfnet:Interpretable multimodal fusion for point cloud registration[J].IEEE Robotics and Automation Letters,2022,7(4):12323-12330.
[19]YEW Z J,LEE J H.Regtr:End-to-end point cloud correspondences with transformers[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:6677-6686.
[20]THOMAS H,QI C R,DESCHAUD J E,et al.Kpconv:Flexible and deformable convolution for point clouds[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:6411-6420.
[21]VASWANI A.Attention is all you need[C]//Advances in Neural Information Processing Systems.2017.
[22]YU H,LI F,SALEH M.Cofinet:Reliable coarse-to-fine corre-spondences for robust point cloud registration[C]//Advances in Neural Information Processing Systems.2021:23872-23884.
[23]LU F,CHEN G,LIU Y,et al.Hregnet:A hierarchical network for large-scale outdoor lidar point cloud registration[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:16014-16023.
[24]QIN Z,YU H,WANG C,et al.Geotransformer:Fast and robust point cloud registration with geometric transformer[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(8):9806-9821.
[25]SINKHORN R,KNOPP P.Concerning nonnegative matricesand doubly stochastic matrices[J].Pacific Journal of Mathema-tics,1967,21(2):343-348.
[26]ARUN K S,HUANG T S,BLOSTEIN S D.Least-squares fitting of two 3-d point sets[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1987,5:698-700.
[27]PASZKE A,GROSS S,MASSA F,et al.Pytorch:An imperative style,high-performance deep learning library[C]//Advances in Neural Information Processing Systems.2019.
[28]KONUR O,KINGMA D,BA J.Adam:A method for stochastic optimization[C]//ICLR.2015:1-15.
[29]ZENG A,SONG S,NIEßNER M,et al.3dmatch:Learning local geometric descriptors from rgb-d reconstructions[C]//Procee-dings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:1802-1811.
[30]GEIGER A,LENZ P,URTASUN R.Are we ready for autonomous driving? the kitti vision benchmark suite[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2012:3354-3361.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发