基于AGConv局部特征描述符的点云配准

doi:10.19678/j.issn.1000-3428.0066359

计算机工程 ›› 2023, Vol. 49 ›› Issue (11): 231-237. doi: 10.19678/j.issn.1000-3428.0066359

基于AGConv局部特征描述符的点云配准

张文丽, 程兰*, 任密蜂, 续欣莹, 阎高伟, 张喆

太原理工大学电气与动力工程学院, 太原 030024

收稿日期:2022-11-25 出版日期:2023-11-15 发布日期:2023-02-08
通讯作者: 程兰
作者简介:
张文丽（1996—），女，硕士研究生，主研方向为三维点云配准
任密蜂，副教授
续欣莹，教授
阎高伟，教授
张喆，讲师
基金资助:
国家自然科学基金(62073232); 国家自然科学基金(61973226); 山西省自然科学基金(201901D211079); 山西省自然科学基金(20210302123189); 山西省科技合作交流基金(202104041101030)

Point Cloud Registration Based on AGConv Local Feature Descriptors

Wenli ZHANG, Lan CHENG*, Mifeng REN, Xinying XU, Gaowei YAN, Zhe ZHANG

College of Electrical and Power Engineering, Taiyuan University of Technology, Taiyuan 030024, China

Received:2022-11-25 Online:2023-11-15 Published:2023-02-08
Contact: Lan CHENG

摘要/Abstract

摘要：

为了提高现有点云配准模型在真实点云数据中的配准精度，基于自适应图卷积(AGConv)的局部特征描述符，提出一种改进的点云配准模型。在数据预处理模块中，通过对点云中的采样点构建局部块并计算局部参考坐标系，规范局部块中的采样点，使其对旋转变换不敏感。在特征提取模块中，利用AGConv为采样点生成自适应核，充分挖掘不同语义部分的点之间的关系，并将规范化的局部块输入基于AGConv的特征提取网络计算局部特征描述符，提高局部特征对遮挡及杂波的鲁棒性。在点云配准模块中，使用随机采样一致性算法估计刚性变换矩阵。在3DMatch数据集上的实验结果表明，相比于DIP模型，该模型的特征匹配和配准召回率分别提高了2.3和5个百分点，能有效提高点云配准精度并且具有较好的鲁棒性。

关键词: 点云配准, 局部块, 局部参考坐标系, 自适应图卷积, 特征描述符

Abstract:

To improve the registration accuracy of existing point cloud registration models for real point cloud data, an improved point cloud registration model is proposed based on the local feature descriptors of Adaptive Graph Convolution(AGConv). In the data preprocessing module, the sampling points in the local patch are normalized by constructing the sampling points in the point cloud and calculating the Local Reference Frame(LRF) to render them insensitive to rotation transformation. In the feature extraction module, AGConv is used to generate an adaptive kernel for the sampling points. The relationships between the points of different semantic parts are fully excavated. The standardized local patches are then input into the AGConv-based feature extraction network to calculate the local feature descriptors and improve the robustness of the local features to occlusion and clutter. In the point cloud registration module, the RANdom SAmpling Consistency(RANSAC) algorithm is used to estimate a rigid transformation matrix. The experimental results on the 3DMatch dataset show that, compared with the DIP model, the Feature Matching Recall(FMR) of this model is increased by 2.3 percentage points, and the Registration Recall(RR) is increased by 5 percentage points. This can effectively improve the registration accuracy of point clouds with good robustness.

Key words: point cloud registration, local patch, Local Reference Frame(LRF), Adaptive Graph Convolution(AGConv), feature descriptor

张文丽, 程兰, 任密蜂, 续欣莹, 阎高伟, 张喆. 基于AGConv局部特征描述符的点云配准[J]. 计算机工程, 2023, 49(11): 231-237.

Wenli ZHANG, Lan CHENG, Mifeng REN, Xinying XU, Gaowei YAN, Zhe ZHANG. Point Cloud Registration Based on AGConv Local Feature Descriptors[J]. Computer Engineering, 2023, 49(11): 231-237.

http://www.ecice06.com/CN/Y2023/V49/I11/231

图/表 8

图1 基于AGConv局部特征描述符的点云配准模型结构

Fig.1 Structure of point cloud registration model based on AGConv local feature descriptors

图2 数据预处理模块结构

Fig.2 Structure of data preprocessing module

图3 特征提取模块结构

Fig.3 Structure of feature extraction module

图4 不同模型的鲁棒性对比

Fig.4 Robustness comparison of different models

图5 配准结果可视化

Fig.5 Visualization of registration results

参考文献 26

1	王忠杰. 结构光三维重建技术研究与应用[D]. 成都: 电子科技大学, 2022.
	WANG Z J. Research and application of structured light 3D reconstruction technology[D]. Chengdu: University of Electronic Science and Technology of China, 2022. (in Chinese)
2	ISHIHARA K, KANERVISTO A, MIURA J, et al. Multi-task learning with attention for end-to-end autonomous driving[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Washington D. C., USA: IEEE Press, 2021: 2896-2905.
3	崔云轩, 刘桂华, 余东应, 等. 点线特征融合的激光雷达单目惯导SLAM系统. 计算机工程, 2022, 48(7): 254- 263. URL
	CUI Y X, LIU G H, YU D Y, et al. Lidar-mono-inertial SLAM system with fusion of point-line features. Computer Engineering, 2022, 48(7): 254- 263. URL
4	ZENG A, SONG S R, NIEßNER M, et al. 3DMatch: learning local geometric descriptors from RGB-D reconstructions[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2017: 199-208.
5	SU H, MAJI S, KALOGERAKIS E, et al. Multi-view convolutional neural networks for 3D shape recognition[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2016: 945-953.
6	PUJOL-MIRÓ A, CASAS J R, RUIZ-HIDALGO J. Correspondence matching in unorganized 3D point clouds using convolutional neural networks. Image and Vision Computing, 2019, 83/84, 51- 60. doi: 10.1016/j.imavis.2019.02.013
7	LI L, ZHU S Y, FU H B, et al. End-to-end learning local multi-view descriptors for 3D point clouds[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 1916-1925.
8	CHARLES R Q, HAO S, MO K C, et al. PointNet: deep learning on point sets for 3D classification and segmentation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2017: 77-85.
9	李建微, 占家旺. 三维点云配准方法研究进展. 中国图象图形学报, 2022, 27(2): 349- 367. URL
	LI J W, ZHAN J W. Review on 3D point cloud registration method. Journal of Image and Graphics, 2022, 27(2): 349- 367. URL
10	AOKI Y, GOFORTH H, SRIVATSAN R A, et al. PointNetLK: robust & efficient point cloud registration using PointNet[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 7156-7165.
11	LUCAS B D, KANADE T. An iterative image registration technique with an application to stereo vision[C]//Proceedings of the 7th International Joint Conference on Artificial Intelligence. New York, USA: ACM Press, 1981: 674-679.
12	WANG Y, SOLOMON J. Deep closest point: learning representations for point cloud registration[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2020: 3522-3531.
13	WANG Y, SUN Y B, LIU Z W, et al. Dynamic graph CNN for learning on point clouds. ACM Transactions on Graphics, 2019, 38(5): 1- 12.
14	WANG Y, SOLOMON J M. PRNet: self-supervised learning for partial-to-partial registration[EB/OL]. [2022-10-11]. https://arxiv.org/abs/1910.12240.
15	FISCHLER M, BOLLES R. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Communications of the Association for Computing Machinery, 1981, 24(6): 381- 395. doi: 10.1145/358669.358692
16	DENG H W, BIRDAL T, ILIC S. PPFNet: global context aware local features for robust 3D point matching[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 195-205.
17	CHOY C, PARK J, KOLTUN V. Fully convolutional geometric features[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2020: 8957-8965.
18	CHOY C, GWAK J, SAVARESE S. 4D spatio-temporal ConvNets: Minkowski convolutional neural networks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 3070-3079.
19	POIESI F, BOSCAINI D. Distinctive 3D local deep descriptors[C]//Proceedings of the 25th International Conference on Pattern Recognition. Washington D. C., USA: IEEE Press, 2021: 5720-5727.
20	GOJCIC Z, ZHOU C F, WEGNER J D, et al. The perfect match: 3D point cloud matching with smoothed densities[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 5540-5549.
21	BAI X Y, LUO Z X, ZHOU L, et al. D3Feat: joint learning of dense detection and description of 3D local features[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 6358-6366.
22	ZHOU H R, FENG Y D, FANG M S, et al. Adaptive graph convolution for point cloud analysis[EB/OL]. [2022-10-11]. https://arxiv.org/abs/2108.08035.
23	KAMOUSI P, LAZARD S, MAHESHWARI A, et al. Analysis of farthest point sampling for approximating geodesics in a graph. Computational Geometry, 2016, 57, 1- 7. doi: 10.1016/j.comgeo.2016.05.005
24	FERNANDEZ-LABRADOR C, CHHATKULI A, PAUDEL D P, et al. Unsupervised learning of category-specific symmetric 3D keypoints from point sets[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2020: 546-563.
25	RUSU R B, BLODOW N, BEETZ M. Fast Point Feature Histograms(FPFH) for 3D registration[C]//Proceedings of IEEE International Conference on Robotics and Automation. Washington D. C., USA: IEEE Press, 2009: 3212-3217.
26	DENG H W, BIRDAL T, ILIC S. PPF-FoldNet: unsupervised learning of rotation invariant 3D local descriptors[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 620-638.

[1]	徐敏,莫东鸣. Canny边缘特征18维描述符在图像拼接中的应用[J]. 计算机工程, 2017, 43(9): 310-315.
[2]	李羿辰,耿国华,张雨禾,李姗姗. 基于区域扩张的三维点云模型配准算法[J]. 计算机工程, 2017, 43(11): 204-209,215.
[3]	周红志，程向阳. 一种基于局部时空特征的视频异常检测方案[J]. 计算机工程, 2014, 40(4): 203-208,213.
[4]	傅红普, 邹北骥. 方向梯度直方图及其扩展[J]. 计算机工程, 2013, 39(5): 212-217.
[5]	赵垒, 侯振杰. 一种改进的SIFT图像配准方法[J]. 计算机工程, 2010, 36(12): 226-228.

选择文件类型/文献管理软件名称

选择包含的内容

基于AGConv局部特征描述符的点云配准

Point Cloud Registration Based on AGConv Local Feature Descriptors

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 26

相关文章 5

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于AGConv局部特征描述符的点云配准

Point Cloud Registration Based on AGConv Local Feature Descriptors

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 26

相关文章 5

编辑推荐

Metrics

本文评价