Survey of 3-Dimensional Point Cloud Processing Based on Deep Learning

被引：0
作者：
Li J. [1 ]
Sun H. [1 ]
Dong Y. [1 ]
Zhang R. [1 ]
Sun X. [1 ,2 ]
机构：
[1] Institute of Computer System, School of Computer and Information Technology, Liaoning Normal University, Dalian
[2] Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia, Beijing University of Posts and Telecommunications, Beijing
来源：
Jisuanji Yanjiu yu Fazhan/Computer Research and Development | 2022年 / 59卷 / 05期
基金：
中国国家自然科学基金;
关键词：
Classification and segmentation; Deep learning; Detection and tracking; Point cloud; Pose estimation; Reconstruction;
D O I：
10.7544/issn1000-1239.20210131
中图分类号：
学科分类号：
摘要：
Deep learning has shown its superior performance in the structured data analysis such as 2-dimensional images. In recent years, with the development of LIDAR sensing equipment and related technologies, 3-dimensional point cloud scanning and acquisition has become more convenient. That makes the analysis and processing of unstructured point cloud data potential become an important research direction and obtain some progress in many fields such as computer graphics, robot, autonomous driving, virtual and augmented reality. A survey on the research of 3-dimensional point cloud processing of recent years is presented. Focusing on the application of deep learning in 3-dimensional point cloud shape analysis, structure extraction, detection and repair, we introduce the extraction method of point cloud topological structure, and compare the progress of the following research directions with the construction of neural networks as the main method: shape deformation, reconstruction, segmentation, classification, object tracking, scene flow estimation, object detection and pose estimation. Finally, we summarize the commonly used 3-dimensional point cloud public datasets, analyze and compare the characteristics and evaluation indicators of various point cloud processing task methods, and point out their advantages and disadvantages. The challenges and development directions of processing point cloud data based on deep learning are discussed. © 2022, Science Press. All right reserved.
引用
收藏
页码：1160 / 1179
页数：19
相关论文
共 100 条
[41]  
Wang Weiyue, Ceylan D, Mech R, Et al., 3DN: 3D deformation network, Proc of the 32nd IEEE/CVF Conf on Computer Vision and Pattern Recognition, pp. 1038-1046, (2019)
[42]  
Zhou Baoxing, Shan Yuhui, Yao Shoufeng, Deformation monitoring of similar material model based on point cloud feature extraction, The Journal of Engineering, 2019, 20, pp. 6790-6794, (2019)
[43]  
Qi C R, Su Hao, Niessner M, Et al., Volumetric and multi-view CNNs for object classification on 3D data, Proc of the 29th IEEE Conf on Computer Vision and Pattern Recognition, pp. 5648-5656, (2016)
[44]  
Tchapmi L, Choy C, Armeni I, Et al., SEGCloud: Semantic segmentation of 3D point clouds, Proc of the 5th Int Conf on 3D Vision, pp. 537-547, (2017)
[45]  
Wang Pengshuai, Liu Yang, Guo Yuxiao, Et al., O-CNN: Octree-based convolutional neural networks for 3D shape analysis, ACM Transactions on Graphics, 36, 4, pp. 72:1-72:11, (2017)
[46]  
Graham B, Engelcke M, Maaten L V D., 3D semantic segmentation with submanifold sparse convolutional networks, Proc of the 31st IEEE/CVF Conf on Computer Vision and Pattern Recognition, pp. 9224-9232, (2018)
[47]  
Meng H, Gao Lin, Lai Yukun, Et al., VV-Net: Voxel vae net with group convolutions for point cloud segmentation, Proc of the 17th IEEE/CVF Int Conf on Computer Vision, pp. 8499-8507, (2019)
[48]  
Shao Tianjia, Yang Yin, Weng Yanlin, Et al., H-CNN: Spatial hashing based CNN for 3D shape analysis, IEEE Transactions on Visualization and Computer Graphics, 26, 7, pp. 2403-2416, (2020)
[49]  
Lefebvre S, Hoppe H., Perfect spatial hashing, ACM Transactions on Graphics, 25, 3, pp. 579-588, (2006)
[50]  
Lawin F J, Danelljan M, Tosteberg P, Et al., Deep projective 3D semantic segmentation, Proc of the 17th Int Conf on Computer Analysis of Images and Patterns, pp. 95-107, (2017)
12345678910