Semantic Segmentation of Point Clouds Using Local Geometric Features and Dilated Neighborhoods

XIANG Xueyong; LI Guangyun; WANG Li; ZONG Wenpeng; LÜ Zhipeng; XIANG Fengzhuo

doi:10.13203/j.whugis20200567

Volume 48 Issue 4

Apr. 2023

Turn off MathJax

Article Contents

Abstract

References

Geomatics and Information Science of Wuhan University > 2023 > 48(4): 534-541. > DOI: 10.13203/j.whugis20200567

XIANG Xueyong, LI Guangyun, WANG Li, ZONG Wenpeng, LÜ Zhipeng, XIANG Fengzhuo. Semantic Segmentation of Point Clouds Using Local Geometric Features and Dilated Neighborhoods[J]. Geomatics and Information Science of Wuhan University, 2023, 48(4): 534-541. DOI: 10.13203/j.whugis20200567

Citation:

PDF (11443 KB)

Semantic Segmentation of Point Clouds Using Local Geometric Features and Dilated Neighborhoods

1.
Institute of Geospatial Information, Information Engineering University, Zhengzhou 450000, China
2.
Xi'an Institute of Surveying and Mapping, Xi'an 710000, China
3.
Troops 61287, Chengdu 610036, China

More Information

Received Date: October 26, 2020
Available Online: April 16, 2023
Published Date: April 04, 2023

Graphical Abstract

Abstract

Abstract

Objectives Point cloud has no topological structure, current deep learning semantic segmentation algorithm is difficult to capture geometric features implied in irregular points. In addition, the point cloud is in three-dimensional space with a large amount of data size. If we blindly expand the captive field size during extract neighborhood information, it will increase the number of model parameters, which will make model training difficult.
Methods We propose a point cloud semantic segmentation model based on the dilated convolution and combining elementary geometric features such as angle as the model input. First, during feature extraction, basic geometric features such as the relative coordinates, distance and angle between the centroid and the neighboring points are used as the model input to mine the geometric information. Second, in the process of building local neighborhoods, we expand the image dilated convolution operator to point cloud processing, the point cloud dilated operator can expand the receptive field size with no increasing the number of parameters of the model. Finally, the dilated convolution operator, multi-geometric features encoding modules and U-Net architecture are combined to form a complete point cloud semantic segmentation model.
Results The results show that compared with the traditional neighborhood structure, the overall accuracy (OA) of dilated neighborhood structure is increased by 1.4%. Compared with the model that only uses coordinates as input, multi-geometric features encoding module is increased by 10.7%. The final model based on the two proposed algorithms get mean intersection over union and OA are 91.2% and 68.2%, respectively.
Conclusions The dilated neighborhood structure can effectively extract point cloud information in a larger range without increasing the number of model parameters. multi-geometric features encoding module can maximize the capture of shape information in the neighborhood.
- point cloud,
- semantic segmentation,
- dilated neighborhood,
- geometric features

FullText(HTML)

References (24)

References

[1]	Hackel T, Wegner J D, Schindler K. Fast Semantic Segmentation of 3D Point Clouds with Strongly Varying Density[J]. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016, Ⅲ-3: 177-184.
[2]	杨必胜, 董震. 点云智能处理[M]. 北京: 科学出版社, 2020. Yang Bisheng, Dong Zhen. Intelligent Processing of Point Cloud[M]. Beijing: Science Press, 2020.
[3]	杨必胜, 梁福逊, 黄荣刚. 三维激光扫描点云数据处理研究进展、挑战与趋势[J]. 测绘学报, 2017, 46(10): 1509-1516. https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201710034.htm Yang Bisheng, Liang Fuxun, Huang Ronggang. Progress, Challenges and Perspectives of 3D LiDAR Point Cloud Processing[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(10): 1509-1516. https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201710034.htm
[4]	Castillo E, Liang J, Zhao H K. Point Cloud Segmentation and Denoising via Constrained Nonlinear Least Squares Normal Estimates[M]//Heidelberg, Berlin: Springer, 2012.
[5]	Anh V. Octree-Based Region Growing for Point Cloud Segmentation[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 104: 88-100. doi: 10.1016/j.isprsjprs.2015.01.011
[6]	张良培, 张云, 陈震中, 等. 基于分裂合并的多模型拟合方法在点云分割中的应用[J]. 测绘学报, 2018, 47(6): 833-843. https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201806017.htm Zhang Liangpei, Zhang Yun, Chen Zhenzhong, et al. Splitting and Merging Based Multi-model Fitting for Point Cloud Segmentation[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(6): 833-843. https://www.cnki.com.cn/Article/CJFDTOTAL-CHXB201806017.htm
[7]	闫利, 谢洪, 胡晓斌, 等. 一种新的点云平面混合分割方法[J]. 武汉大学学报(信息科学版), 2013, 38(5): 517-521. http://ch.whu.edu.cn/article/id/2631 Yan Li, Xie Hong, Hu Xiaobin, et al. A New Hybrid Plane Segmentation Approach of Point Cloud[J]. Geomatics and Information Science of Wuhan University, 2013, 38(5): 517-521. http://ch.whu.edu.cn/article/id/2631
[8]	熊汉江, 郑先伟, 丁友丽, 等. 基于2D-3D语义传递的室内三维点云模型语义分割[J]. 武汉大学学报(信息科学版), 2018, 43(12): 2303-2309. doi: 10.13203/j.whugis20180190 Xiong Hanjiang, Zheng Xianwei, Ding Youli, et al. Semantic Segmentation of Indoor 3D Point Cloud Model Based on 2D-3D Semantic Transfer[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 2303-2309. doi: 10.13203/j.whugis20180190
[9]	Milioto A, Vizzo I, Behley J, et al. RangeNet: Fast and Accurate LiDAR Semantic Segmentation[C]// IEEE International Conference on Intelligent Robots and Systems (IROS), Macau, China, 2020.
[10]	Maturana D, Scherer S. VoxNet: A 3D Convolutional Neural Network for Real-Time Object Recognition[C]// IEEE International Conference on Intelligent Robots and Systems (IROS), Hamburg, Germany, 2015.
[11]	Charles R Q, Hao S, Mo K C, et al. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, 2017.
[12]	Qi C R, Yi L, Su H, et al. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space[C]//The 31st International Conference on Neural Information Processing Systems, Long Beach, California, USA, 2017.
[13]	Fisher Y, Koltun V. Multi-scale Context Aggregation by Dilated Convolutions[C]//International Conference on Learning Representations, San Juan, Puerto Rico, USA, 2016.
[14]	Jiang M Y, Wu Y R, Zhao T Q, et al. PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation[J]. arXiv, 2018, DOI: 1807.00652.
[15]	Li Y, Bu R, Sun M, et al. PointCNN: Convolution on X-Transformed Points[C]. Advances in Neural Information Processing Systems, Montréal, Canada, 2018.
[16]	Liang Z D, Yang M, Li H, et al. 3D Instance Embedding Learning with a Structure-Aware Loss Function for Point Cloud Segmentation[J]. IEEE Robotics and Automation Letters, 2020, 5(3): 4915-4922.
[17]	Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation[M]// Cham: Springer International Publishing, 2015.
[18]	Zhao H S, Jiang L, Fu C W, et al. PointWeb: Enhancing Local Neighborhood Features for Point Cloud Processing[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, USA, 2020.
[19]	Thomas H, Qi C R, Deschaud J E, et al. KPConv: Flexible and Deformable Convolution for Point Clouds[C]// IEEE International Conference on Computer Vision (ICCV), Seoul, Korea (South), 2020.
[20]	Hackel T, Savinov N, Ladicky L, et al. Semantic3D. Net: A New Large-Scale Point Cloud Classification Benchmark[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, IV(1): 91-98.
[21]	Wang F, Zhuang Y, Gu H, et al. OctreeNet: A Novel Sparse 3-D Convolutional Neural Network for Real-Time 3-D Outdoor Scene Analysis[J]. IEEE Transactions on Automation Science and Engineering, 2020, 17(2): 735-747.
[22]	Contreras J, Denzler J. Edge-convolution Point Net for Semantic Segmentation of Large-scale Point Clouds[C]// IEEE International Geoscience and Remote Sensing Symposium, Yokohama, Japan, 2019.
[23]	Thomas H, Goulette F, Deschaud J E, et al. Semantic Classification of 3D Point Clouds with Multiscale Spherical Neighborhoods[C]// International Conference on 3D Vision (3DV), Verona, Italy, 2018.
[24]	Roynard X, Deschaud J E, Goulette F. Classification of Point Cloud Scenes with Multiscale Voxel Deep Network[J]. arXiv, 2018, DOI: 1804.03583.

[1]	XIE Yakun, ZHANG Yang, HU Yunong, ZHAN Ni, SUN Ting, ZHU Jun, ZHU Qing. A Lightweight Approach to Railway Infrastructure BIM Models Considering Geometric Detail Features[J]. Geomatics and Information Science of Wuhan University. DOI: 10.13203/j.whugis20240254
[2]	MA Tian-en, LIU Tao, DU Ping, CHEN Po-yi, LING Zhen-fei. A 3D Point Cloud Semantic Segmentation Method for Aggregating Global Context Information[J]. Geomatics and Information Science of Wuhan University. DOI: 10.13203/j.whugis20230143
[3]	LI Qingquan, WANG Chisheng, XIONG Siting, ZHANG Dejin, ZOU Qin, TU Wei. Generalized Surveying Data Processing: From Geometric Parameters Calculation to Feature Information Extraction[J]. Geomatics and Information Science of Wuhan University, 2022, 47(11): 1805-1814. DOI: 10.13203/j.whugis20220606
[4]	JIANG Tengping, YANG Bisheng, ZHOU Yuzhou, ZHU Runsong, HU Zongtian, DONG Zhen. Bilevel Convolutional Neural Networks for 3D Semantic Segmentation Using Large-scale LiDAR Point Clouds in Complex Environments[J]. Geomatics and Information Science of Wuhan University, 2020, 45(12): 1942-1948. DOI: 10.13203/j.whugis20200081
[5]	LUMiao, MEI Yang, ZHAO Yong, LENG Liang. Change Detection Based on Multi-scale Geometric Feature Vector[J]. Geomatics and Information Science of Wuhan University, 2015, 40(5): 623-627. DOI: 10.13203/j.whugis20130382
[6]	Pang Shiyan, Liu YawenZuo Zhiqi, Chen Zhengfu, . Combination of Region Growing and TIN Edge Segmentation for Extraction of Geometric Features on Building Facades[J]. Geomatics and Information Science of Wuhan University, 2015, 40(1): 102-106.
[7]	WANG Huibing, TANG Xinming, QIU Bo, WANG Wenjie. Geometric Matching Method of Area Feature Based on Multi-weighted Operators[J]. Geomatics and Information Science of Wuhan University, 2013, 38(10): 1243-1247.
[8]	SUN Yizhong, YAO Chi, CHEN Shaoqin, XU Wenxiang. Geographical Elements′ Spatial Location Identification Considering Geometric Features[J]. Geomatics and Information Science of Wuhan University, 2012, 37(12): 1486-1489.
[9]	WU Zhaocong, QIN Maoyun, ZHANG Xiao. A Cloud-Model-Based Remote Sensing Image Segmentation Concerned with Geometrical Features[J]. Geomatics and Information Science of Wuhan University, 2008, 33(9): 939-942.
[10]	Shi Wenzhong. Error Models for Geometric Features in Three Dimensional GIS[J]. Geomatics and Information Science of Wuhan University, 1998, 23(1): 18-20,25.

Cited By

Cited by

Periodical cited type(4)

1.	武斌，刘溢安，赵洁. 结合空间结构卷积和注意力机制的三维点云分类网络. 中国图象图形学报. 2024(02): 520-532 .
2.	张秋昭，梁嘉辉，段浩然，王宗伟，段伟. 基于空间几何特征融合增强的地铁隧道点云语义分割神经网络模型. 金属矿山. 2023(05): 237-246 .
3.	吴立韬，孙啸，田正宏. 渠道预制块自动化铺装机械检测定位技术研究. 水力发电. 2023(12): 67-74 .
4.	何春秀，荆现文，何永宁. 分组自注意力机制的多层级三维点云分类方法. 计算机工程与应用. 2023(24): 259-267 .

Other cited types(2)

Get Citation

PDF

XML

Article views (1371) PDF downloads (132) Cited by(6)

Semantic Segmentation of Point Clouds Using Local Geometric Features and Dilated Neighborhoods

Abstract

References

Related Articles

Cited by

Periodical cited type(4)

Other cited types(2)

Catalog

Related

Semantic Segmentation of Point Clouds Using Local Geometric Features and Dilated Neighborhoods

Abstract

References

Related Articles

Cited by

Periodical cited type(4)

Other cited types(2)

Catalog

Related

Export File

Citation

Format

Content