PatchCNN: An Explicit Convolution Operator for Point Clouds Perception
文献类型:期刊论文
作者 | Wang F(王斐)2; Zhang, Xing2; Jiang Y(姜勇)1![]() |
刊名 | IEEE Geoscience and Remote Sensing Letters
![]() |
出版日期 | 2021 |
卷号 | 18期号:4页码:726-730 |
关键词 | Deep learning explicit convolution geometric relationship point cloud perception |
ISSN号 | 1545-598X |
产权排序 | 2 |
英文摘要 | A novel convolution architecture PatchCNN is proposed for extending 2-D grid convolution to the nongrid structured data: point clouds, without any intermediate data representation. Previous studies implicitly capture local shape pattern from the meaningful subset or a local region without considering the interaction among points of the local region. The PointPatch module in our deep network, in spirit to the 8-pixels neighborhood in the 2-D image, explicitly models geometric relationship among points in the local region. We adopt a light 3-D convolution network to adaptively integrate features of the PointPatch module. The integrated features encode geometric relationship and the impact of surrounding points, which brings sufficient shape awareness and robustness for point cloud perception. Additionally, in our work, the convolution weight on each point is treated as a Lipschitz continuous function approximated by multilayer perceptron (MLP) and integrated features in the local region. Theoretically, the explicit learning strategy proposed in PatchCNN introduces inductive bias beneficial to the learning shape pattern in 3-D Euclidean space. Extensive experiments on ModelNet40 and ScanNet v2 data sets demonstrate that the proposed method achieves the competitive performance on par or even better than state-of-The-Art methods. © 2004-2012 IEEE. |
资助项目 | Fundamental Research Funds for the Central Universities of China[N172608005] ; Fundamental Research Funds for the Central Universities of China[N182612002] ; Fundamental Research Funds for the Central Universities of China[N2026002] ; Liaoning Provincial Natural Science Foundation of China[20180520007] |
WOS研究方向 | Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology |
语种 | 英语 |
WOS记录号 | WOS:000633394400033 |
资助机构 | Fundamental Research Funds for the Central Universities of China under Grant N172608005, Grant N182612002, and Grant N2026002 ; Liaoning Provincial Natural Science Foundation of China under Grant 20180520007 |
源URL | [http://ir.sia.cn/handle/173321/28709] ![]() |
专题 | 工艺装备与智能机器人研究室 |
通讯作者 | Zhang, Xing |
作者单位 | 1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China 2.Faculty of Robot Science and Engineering, Northeastern University, Shenyang, China |
推荐引用方式 GB/T 7714 | Wang F,Zhang, Xing,Jiang Y,et al. PatchCNN: An Explicit Convolution Operator for Point Clouds Perception[J]. IEEE Geoscience and Remote Sensing Letters,2021,18(4):726-730. |
APA | Wang F,Zhang, Xing,Jiang Y,Kong, Li,&Wei, Xiaotong.(2021).PatchCNN: An Explicit Convolution Operator for Point Clouds Perception.IEEE Geoscience and Remote Sensing Letters,18(4),726-730. |
MLA | Wang F,et al."PatchCNN: An Explicit Convolution Operator for Point Clouds Perception".IEEE Geoscience and Remote Sensing Letters 18.4(2021):726-730. |
入库方式: OAI收割
来源:沈阳自动化研究所
其他版本
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。