Deep Learning on Point Clouds and Its Application: A Survey

Author:

Liu Weiping,Sun Jia,Li WanyiORCID,Hu TingORCID,Wang Peng

Abstract

Point cloud is a widely used 3D data form, which can be produced by depth sensors, such as Light Detection and Ranging (LIDAR) and RGB-D cameras. Being unordered and irregular, many researchers focused on the feature engineering of the point cloud. Being able to learn complex hierarchical structures, deep learning has achieved great success with images from cameras. Recently, many researchers have adapted it into the applications of the point cloud. In this paper, the recent existing point cloud feature learning methods are classified as point-based and tree-based. The former directly takes the raw point cloud as the input for deep learning. The latter first employs a k-dimensional tree (Kd-tree) structure to represent the point cloud with a regular representation and then feeds these representations into deep learning models. Their advantages and disadvantages are analyzed. The applications related to point cloud feature learning, including 3D object classification, semantic segmentation, and 3D object detection, are introduced, and the datasets and evaluation metrics are also collected. Finally, the future research trend is predicted.

Funder

National Natural Science Foundation of China

China Postdoctoral Science Foundation

Youth Innovation Promotion Association of the Chinese Academy of Sciences

Publisher

MDPI AG

Subject

Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry

Reference119 articles.

1. Generation of Visually Aesthetic and Detailed 3d Models of Historical Cities by Using Laser Scanning and Digital Photogrammetry;Balsabarreiro;Digit. Appl. Archaeol. Cult. Herit.,2017

2. Generation of 3d/4d Photorealistic Building Models. The Testbed Area for 4d Cultural Heritage World Project: The Historical Center of Calw (Germany);Balsa-Barreiro,2015

3. 3D object perception and perceptual learning in the RACE project

4. Learning ambidextrous robot grasping policies

5. Velodyne Hdl-64e Lidar Specificationhttps://velodynelidar.com/hdl-64e.html

Cited by 149 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Deep learning for 3D object recognition: A survey;Neurocomputing;2024-12

2. Geometric characterization and segmentation of historic buildings using classification algorithms and convolutional networks in HBIM;Automation in Construction;2024-11

3. Synthesizing realistic sand assemblies with denoising diffusion in latent space;International Journal for Numerical and Analytical Methods in Geomechanics;2024-08-14

4. Point cloud semantic segmentation of grotto scenes using the knowledge-guided deep learning method;International Journal of Digital Earth;2024-08-04

5. Sequential Point Clouds: A Survey;IEEE Transactions on Pattern Analysis and Machine Intelligence;2024-08

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3