Institutional Repository of Coll Elect & Informat Engn
A Review of Point Cloud 3D Object Detection Methods Based on Deep Learning | |
Wang, Xiyuan; Lin, Jie![]() | |
2024 | |
会议名称 | 38th CCF National Conference of Computer Applications, CCF NCCA 2023 |
会议录名称 | Communications in Computer and Information Science
![]() |
卷号 | 1959 CCIS |
页码 | 30-39 |
会议日期 | July 16, 2023 - July 20, 2023 |
会议地点 | Suzhou, China |
出版者 | Springer Science and Business Media Deutschland GmbH |
摘要 | Based on introducing the coupling relationship between deep learning and three-dimensional point clouds, this paper reviews the three characteristics and research problems of point clouds, randomness, sparsity, and unstructuredness, and discusses three-dimensional point cloud target detection based on deep neural networks, including point cloud detection techniques following graph convolution, detection techniques following the original point cloud, and detection algorithms based on fusion processing of graph convolution and the original point cloud. Focusing on future research direction and development, the field of point cloud analysis is currently undergoing further development through the application of deep learning techniques. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. |
关键词 | Convolutional neural networks - Deep neural networks - Learning systems - Object detection - Object recognition - Semantic Segmentation - Semantics 3D object - 3D point cloud - Coupling relationships - Deep leaning - Object detection method - Point-clouds - Research problems - Semantic segmentation - Targets detection - Three-dimensional point clouds |
DOI | 10.1007/978-981-99-8764-1_3 |
收录类别 | EI |
语种 | 英语 |
EI入藏号 | 20235215288801 |
EI主题词 | Convolution |
EI分类号 | 461.4 Ergonomics and Human Factors Engineering - 716.1 Information Theory and Signal Processing - 723.2 Data Processing and Image Processing - 723.4 Artificial Intelligence |
ISSN | 1865-0929 |
原始文献类型 | Conference article (CA) |
EISSN | 1865-0937 |
引用统计 | 无
|
文献类型 | 会议论文 |
条目标识符 | https://ir.lut.edu.cn/handle/2XXMBERH/169323 |
专题 | 电气工程与信息工程学院 |
通讯作者 | Lin, Jie |
作者单位 | College of Electrical Engineering and Information Engineering, Lanzhou University of Technology, Lanzhou; 730050, China |
第一作者单位 | 兰州理工大学 |
通讯作者单位 | 兰州理工大学 |
推荐引用方式 GB/T 7714 | Wang, Xiyuan,Lin, Jie,Yang, Longrui,et al. A Review of Point Cloud 3D Object Detection Methods Based on Deep Learning[C]:Springer Science and Business Media Deutschland GmbH,2024:30-39. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论