├── Dataset.png ├── README.md ├── Segmentation.pdf ├── Segmentation.png ├── classification.pdf ├── classification.png ├── detection.pdf ├── detection.png ├── detection_bev.pdf ├── detection_bev.png └── taxonomy.png /Dataset.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/Dataset.png -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | [![arXiv](https://img.shields.io/badge/arXiv-1912.12033-b31b1b.svg)](https://arxiv.org/abs/1912.12033) 2 | [![Maintenance](https://img.shields.io/badge/Maintained%3F-yes-green.svg)](https://github.com/QingyongHu/SoTA-Point-Cloud/graphs/commit-activity) 3 | [![GitHub issues](https://img.shields.io/github/issues/QingyongHu/SoTA-Point-Cloud)](https://GitHub.com/QingyongHu/SoTA-Point-Cloud/issues/) 4 | [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg?style=flat-square)](http://makeapullrequest.com) 5 | 6 | 7 | # Deep Learning for 3D Point Clouds: A Survey (IEEE TPAMI, 2020) 8 | 9 | This is the official repository of **Deep Learning for 3D Point Clouds: A Survey** (IEEE TPAMI), a comprehensive survey 10 | of recent progress in deep learning methods for point clouds. For details, please refer to: 11 | 12 | **Deep Learning for 3D Point Clouds: A Survey**
13 | 14 | [Yulan Guo∗](http://yulanguo.me/), 15 | [Hanyun Wang∗](https://scholar.google.com.hk/citations?user=QG3LdUcAAAAJ&hl=zh-CN), 16 | [Qingyong Hu∗](https://qingyonghu.github.io/), Hao Liu∗, 17 | [Li Liu](http://www.ee.oulu.fi/~lili/LiLiuHomepage.html), 18 | and [Mohammed Bennamoun](http://staffhome.ecm.uwa.edu.au/~00051632/).
19 | (* *indicates equal contribution*) 20 | 21 | **[[Paper](https://arxiv.org/abs/1912.12033)] [[Blog](https://mp.weixin.qq.com/s/5RJAv_cOlhee1R9uZzkmHQ)]**
22 | 23 |

24 | 25 | 26 | ## Introduction 27 | We present a comprehensive review of recent deep learning methods for point clouds. It covers major tasks in 3D point cloud analysis, 28 | including 3D shape classification, 3D object detection, and 3D point cloud segmentation. It also presents comparative 29 | results on several publicly available datasets, together with insightful observations and inspiring future research directions. 30 | Please feel free to contact me 31 | or [create an issue](https://help.github.com/en/github/managing-your-work-on-github/creating-an-issue) on this page if you have new results to add or any suggestions! 32 | 33 | We will update this page on a regular basis! So stay tuned~ :tada::tada::tada: 34 | 35 | ### (1) Datasets 36 |

37 | 38 | ### (2) 3D Shape Classification 39 | #### Public Datasets 40 | - ModelNet (CVPR'15) [[paper]](http://3dvision.princeton.edu/projects/2014/3DShapeNets/paper.pdf) [[project page]](http://modelnet.cs.princeton.edu/) 41 | - ModelNet10 [[data]](http://3dvision.princeton.edu/projects/2014/3DShapeNets/ModelNet10.zip) [[results]](http://modelnet.cs.princeton.edu/) 42 | - ModelNet40 [[data]](http://modelnet.cs.princeton.edu/ModelNet40.zip) [[results]](http://modelnet.cs.princeton.edu/) 43 | - PartNet (CVPR'19) [[paper]](https://arxiv.org/abs/1812.02713) [[data]](https://github.com/daerduoCarey/partnet_dataset) [[project page]](https://cs.stanford.edu/~kaichun/partnet/) 44 | - ScanObjectNN (ICCV'19) [[paper]](https://arxiv.org/pdf/1908.04616.pdf) [[data]](https://github.com/hkust-vgd/scanobjectnn) [[project page]](https://hkust-vgd.github.io/scanobjectnn/) 45 | 46 | #### Benchmark Results 47 |

48 | 49 | ### (3) 3D Object Detection 50 | #### Public Datasets 51 | - KITTI (CVPR'12) [[paper]](http://www.cvlibs.net/publications/Geiger2012CVPR.pdf) [[project page]](http://www.cvlibs.net/datasets/kitti/eval_3dobject.php) 52 | - _3D objecct detection_ [[data]](http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d) [[results]](http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d) 53 | - _BEV_ [[data]](http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=bev) [[results]](http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=bev) 54 | - ApolloScape (TPAMI'19) [[paper]](http://ad-apolloscape.bj.bcebos.com/public%2FApolloScape%20Dataset.pdf) [[data]](http://apolloscape.auto/tracking.html#to_data_href) [[results]](http://apolloscape.auto/leader_board.html) 55 | - Argoverse (CVPR'19) [[paper]](http://openaccess.thecvf.com/content_CVPR_2019/papers/Chang_Argoverse_3D_Tracking_and_Forecasting_With_Rich_Maps_CVPR_2019_paper.pdf) [[data]](https://www.argoverse.org/data.html#download-link) [[project page]](https://www.argoverse.org/index.html) 56 | - A*3D (arXiv'19) [[paper]](https://arxiv.org/pdf/1909.07541) [[data]](https://github.com/I2RDL2/ASTAR-3D#Download) [[project page]](https://github.com/I2RDL2/ASTAR-3D) 57 | - Waymo (arXiv'19) [[paper]](https://arxiv.org/pdf/1912.04838) [[data]](https://waymo.com/open/licensing/) [[project page]](https://waymo.com/open/) 58 | 59 | #### Benchmark Results 60 |

61 |

62 | 63 | 64 | ### (4) 3D Point Cloud Segmentation 65 | #### Public Datasets 66 | - Semantic3D (ISPRS'17) [[paper]](https://www.ethz.ch/content/dam/ethz/special-interest/baug/igp/photogrammetry-remote-sensing-dam/documents/pdf/Papers/Hackel-etal-cmrt2017.pdf) [[project page]](http://www.semantic3d.net/) 67 | - _semantic-8_ [[data]](http://www.semantic3d.net/view_dbase.php?chl=1#download) [[results]](http://www.semantic3d.net/view_results.php?chl=1) 68 | - _reduced-8_ [[data]](http://www.semantic3d.net/view_dbase.php?chl=2#download) [[results]](http://www.semantic3d.net/view_results.php?chl=2) 69 | - S3DIS (CVPR'17) [[paper]](http://buildingparser.stanford.edu/images/3D_Semantic_Parsing.pdf) [[data]](https://docs.google.com/forms/d/e/1FAIpQLScDimvNMCGhy_rmBA2gHfDu3naktRm6A8BPwAWWDv-Uhm6Shw/viewform?c=0&w=1) [[project page]](http://buildingparser.stanford.edu/dataset.html#Download) 70 | - ScanNet (CVPR'17) [[paper]](https://arxiv.org/pdf/1702.04405) [[data]](https://github.com/ScanNet/ScanNet) [[project page]](http://www.scan-net.org/) [[results]](http://kaldir.vc.in.tum.de/scannet_benchmark/) 71 | - NPM3D (IJRR'18) [[paper]](https://arxiv.org/pdf/1712.00032) [[data]](https://cloud.mines-paristech.fr/index.php/s/JhIxgyt0ALgRZ1O) [[project page]](http://npm3d.fr/) [[results]](http://npm3d.fr/paris-lille-3d) 72 | - DublinCity (BMVC'19) [[paper]](https://arxiv.org/abs/1909.03613) [[data]](https://v-sense.scss.tcd.ie/dublincity/) [[project page]](https://v-sense.scss.tcd.ie/dublincity/) 73 | - SemanticKITTI (ICCV'19) [[paper]](https://arxiv.org/pdf/1904.01416) [[data]](http://semantic-kitti.org/dataset.html#download) [[project page]](http://semantic-kitti.org/index.html) [[results]](https://competitions.codalab.org/competitions/20331#results) 74 | - nuScenes (CVPR'20) [[paper]](https://arxiv.org/abs/1903.11027) [[data]](https://www.nuscenes.org/lidar-segmentation) [[project page]](https://www.nuscenes.org/lidar-segmentation) [[results]](https://www.nuscenes.org/lidar-segmentation) 75 | - Toronto-3D (CVPRW'20) [[paper]](https://arxiv.org/abs/2003.08284) [[data]](https://github.com/WeikaiTan/Toronto-3D) [[project page]](https://github.com/WeikaiTan/Toronto-3D) [[results]](https://github.com/WeikaiTan/Toronto-3D) 76 | - DALES (CVPRW'20) [[paper]](https://arxiv.org/abs/2004.11985) [[data]](https://docs.google.com/forms/d/e/1FAIpQLSe3IaTxCS7wKH01SHn_o7U86ToIw9K26vc0bkwiELn6wwh8gg/viewform) [[project page]](https://udayton.edu/engineering/research/centers/vision_lab/research/was_data_analysis_and_processing/dale.php) [[results]](https://arxiv.org/abs/2004.11985) 77 | - Campus3D (ACM MM'20) [[paper]](https://arxiv.org/abs/2008.04968) [[data]](https://3d.dataset.site/) [[project page]](https://github.com/shinke-li/Campus3D) [[results]](https://arxiv.org/abs/2008.04968) 78 | - SensatUrban (CVPR'21) [[paper]](https://arxiv.org/abs/2009.03137) [data][[project page]](http://point-cloud-analysis.cs.ox.ac.uk/) [[results]](https://arxiv.org/abs/2009.03137) 79 | 80 | 81 | #### Benchmark Results 82 |

83 | 84 | 85 | ### Citation 86 | If you find our work useful in your research, please consider citing: 87 | 88 | @article{guo2020deep, 89 | title={Deep learning for 3d point clouds: A survey}, 90 | author={Guo, Yulan and Wang, Hanyun and Hu, Qingyong and Liu, Hao and Liu, Li and Bennamoun, Mohammed}, 91 | journal={IEEE transactions on pattern analysis and machine intelligence}, 92 | year={2020}, 93 | publisher={IEEE} 94 | } 95 | 96 | ## Updates 97 | * 26/02/2020: Adding the dataset information 98 | * 27/12/2019: Initial release. 99 | 100 | 101 | ## Related Repos 102 | 1. [RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds](https://github.com/QingyongHu/RandLA-Net) ![GitHub stars](https://img.shields.io/github/stars/QingyongHu/RandLA-Net.svg?style=flat&label=Star) 103 | 2. [SensatUrban: Learning Semantics from Urban-Scale Photogrammetric Point Clouds](https://github.com/QingyongHu/SensatUrban) ![GitHub stars](https://img.shields.io/github/stars/QingyongHu/SensatUrban.svg?style=flat&label=Star) 104 | 3. [3D-BoNet: Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds](https://github.com/Yang7879/3D-BoNet) ![GitHub stars](https://img.shields.io/github/stars/Yang7879/3D-BoNet.svg?style=flat&label=Star) 105 | 4. [SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration](https://github.com/QingyongHu/SpinNet) ![GitHub stars](https://img.shields.io/github/stars/QingyongHu/SpinNet.svg?style=flat&label=Star) 106 | 5. [SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds with 1000x Fewer Labels](https://github.com/QingyongHu/SQN) ![GitHub stars](https://img.shields.io/github/stars/QingyongHu/SQN.svg?style=flat&label=Star) 107 | 108 | 109 | 110 | -------------------------------------------------------------------------------- /Segmentation.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/Segmentation.pdf -------------------------------------------------------------------------------- /Segmentation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/Segmentation.png -------------------------------------------------------------------------------- /classification.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/classification.pdf -------------------------------------------------------------------------------- /classification.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/classification.png -------------------------------------------------------------------------------- /detection.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/detection.pdf -------------------------------------------------------------------------------- /detection.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/detection.png -------------------------------------------------------------------------------- /detection_bev.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/detection_bev.pdf -------------------------------------------------------------------------------- /detection_bev.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/detection_bev.png -------------------------------------------------------------------------------- /taxonomy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/QingyongHu/SoTA-Point-Cloud/7e9db6b05a5cd643bc24ee0252f898e133032a2f/taxonomy.png --------------------------------------------------------------------------------