点云配准的核心是估算变换矩阵.两个点云对之间存在部分重叠、高噪声和密度差异,现有方法无法精确解决显著点云对应关系之间特征对齐问题.因此,提出显著性峰值与特征对齐网络(significance peak and feature alignment network, SPFANet),实现由粗到细的点云配准.SPFANet由多显著性峰值检测器、粗配准和细配准三部分组成.首先,多显著性峰值检测器引入一种基于描述符方差和重叠分数重加权峰值损失方法,去除非歧视与非重叠的关键点云;其次,粗配准阶段通过检测互补关键点集来计算粗步的配准方案;最后,细配准阶段引入带有前向后向变换的特征度量框架细化粗配准,完成高效点云配准.在同源3DMatch数据集和跨源3DCSR数据集上的实验验证了SPFANet的有效性.
Abstract
The core of point cloud registration is to estimate the transformation matrix. There is partial overlap, high noise, and density difference between two point cloud pairs. The existing methods cannot accurately solve the problem of feature alignment between significant point cloud correspondences. Therefore, a significance peak and feature alignment network (SPFANet) is proposed to achieve point cloud registration from coarse to fine one. SPFANet consists of three parts: multi-significance peak detector, coarse registration, and fine registration. Firstly, the multi-significance peak detector introduces a re-weighted peak loss method based on descriptor variance and overlap score to remove non discriminatory and non overlapping key point clouds. Secondly, the coarse registration stage detects the complementary key point sets to compute the coarse registration scheme. Finally, the fine registration stage introduces a feature metric framework with a forward-backward transform to refine the coarse registration scheme and achieve efficient point cloud registration. The effectiveness of SPFANet is validated through experiments on the same source 3DMatch dataset and cross-source 3DCSR dataset.
BESLP J, MCKAYN D .A method for registration of 3-D shapes[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence. IEEE,1992:239-256.
[2]
YANGJ, LIH, CAMPBELLD, et al. Go-ICP : a globally optimal solution to 3D ICP point-set registration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016,38(11):2241-2254.
[3]
JIANB, VEMURIB C. Robust point set registration using Gaussian mixture models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8):1633-1645.
[4]
HUANGX S, ZHANGJ, FANL X,et al .A systematic approach for cross-source point cloud registration by preserving macro and micro structures[J].IEEE Transactions on Image Processing,2017, 26(7): 3261-3276.
[5]
HUANGX S, FANL X, WUQ,et al .Fast registration for cross-source point clouds by using weak regional affinity and pixel-wise refinement[C]//2019 IEEE International Conference on Multimedia and Expo (ICME). Shanghai,China. IEEE,2019:1552-1557.
[6]
FISCHLERM A, BOLLESR C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6): 381-395.
[7]
MELLADON, AIGERD, MITRAN J,et al .Super 4PCS fast global pointcloud registration via smart indexing[C]//Proceedings of the Symposium on Geometry Processing. ACM,2014: 205-215.
[8]
STEDERB, RUSUR B, KONOLIGEK, et al. NARF: 3D range image features for object recognition[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS Workshop. 2010:44.
[9]
MIANA, BENNAMOUNM, OWENSR .On the repeatability and quality of keypoints for local feature-based 3D object retrieval from cluttered scenes[J]. International Journal of Computer Vision, 2010, 89(2): 348-361.
[10]
ZHONGY. Intrinsic shape signatures:a shape descriptor for 3D object recognition[C]//2009 IEEE 12th International Conference on Computer Vision Workshops,ICCV Workshops. Kyoto, Japan. IEEE, 2009: 689-696.
[11]
RUSUR B, BLODOWN, MARTONZ C,et al .Aligning point cloud views using persistent feature histograms[C]//2008 IEEE/RSJ International Conference on Intelligent Robots and Systems. Nice,France. IEEE,2008:3384-3391.
[12]
RUSUR B, BLODOWN, BEETZM,et al .Fast point feature histograms (FPFH) for 3D registration[C]//Proceedings of the 2009 IEEE International Conference on Robotics and Automation. Kobe, Japan. ACM, 2009: 1848-1853.
[13]
ZHOUQ Y, PARKJ, KOLTONV. Fast global registration[C]//Computer Vision-ECCV 2016: 14th European Conference. Amsterdam, The Netherlands. 2016: 766-782.
[14]
ZENGA, SONGS R, NIEßNERM,et al .3DMatch:learning local geometric descriptors from RGB-D reconstructions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu,HI,USA. IEEE,2017:199-208.
[15]
BAIX Y, LUOZ X, ZHOUL, et al. D3Feat: joint learning of dense detection and description of 3D local features[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle,WA,USA. IEEE,2020: 6358-6366.
LUJ, CHENK, FANZ J. Point cloud registration algorithm based on scale difference descriptor of point neighborhood[J]. Journal of National University of Defense Technology,2021,43(3): 128-134.(in Chinese)
XIONGF W, ZHUANGJ, SHENR. Point cloud with low feature recognition registration method based on neighborhood curvature[J]. Application Research of Computers,2022,39(1):285-290.(in Chinese)
[22]
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). Long Beach, CA, USA. IEEE, 2019:7163-7172.
[23]
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). Seattle,WA,USA. IEEE,2020:11363-11371.
[24]
LIX Q, PONTESJ K, LUCEYS .PointNetLK revisited[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville,TN,USA. IEEE,2021.
HUANGS Y, GOJCICZ, USVYATSOVM,et al. Predator: registration of 3D point clouds with low overlap[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville,TN,USA. IEEE,2021:4265-4274.
[31]
LIJ H, ZHANGC H, XUZ Y,et al .Iterative distance-aware similarity matrix convolution with mutual-supervised point elimination for efficient point cloud registration[C]// Computer Vision-ECCV 2020.Cham:Springer International Publishing,2020: 378-394.
[32]
YEWZ J, LEEG H. REGTR:end-to-end point cloud correspondences with transformers[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans,LA,USA. IEEE,2022:6667-6676.
[33]
QINZ, YUH, WANGC, et al. Geometric transformer for fast and robust point cloud registration[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans,LA,USA. IEEE,2022:11133-11142.
[34]
CHENGX, YUH, ZHANGH, et al. WeightedPose: generalizable cross-pose estimation via weighted SVD[J].arXiv, 2024, arXiv:
[35]
BAKERS, MATTHEWSI .Lucas-kanade 20 years on:a unifying framework[J].International Journal of Computer Vision,2004,56(3): 221-255.
[36]
GROUEIXT, FISHERM, KIMV G,et al. A papier-Mache approach to learning 3D surface generation[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA. IEEE, 2018: 216-224.
[37]
HUANGX S, MEIG F, ZHANGJ, et al. A comprehensive survey on point cloud registration[EB/OL]. [2024-02-18]
LEEJ, KIMS, CHOM, et al. Deep Hough voting for robust global registration[C]// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. IEEE, 2021:15974-15983.
[40]
CHOYC, DONGW, KOLTUNV. Deep global registration[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 2514-2523.
[41]
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). Montreal,QC,Canada. IEEE,2021: 3112-3121.
[42]
MELLADON, DELLEPIANEM, SCOPIGNOR .Relative scale estimation and 3D registration of multi-modal geometry using growing least squares[J].IEEE Transactions on Visualization and Computer Graphics,2016,22(9):2160-2173.
[43]
HUANGX S, ZHANGJ, WUQ,et al .A coarse-to-fine algorithm for registration in 3D street-view cross-source point clouds[C]//2016 International Conference on Digital Image Computing:Techniques and Applications (DICTA). Gold Coast,QLD,Australia. IEEE,2016:1-6.
[44]
PENGF R, WUQ, FANL X,et al .Street view cross-sourced point cloud matching and registration[C]//2014 IEEE International Conference on Image Processing (ICIP). Paris,France. IEEE, 2014:2026-2030.