{"id":15626368,"url":"https://github.com/huangcongqing/3d-point-clouds","last_synced_at":"2025-04-08T08:17:49.915Z","repository":{"id":43894620,"uuid":"329868761","full_name":"HuangCongQing/3D-Point-Clouds","owner":"HuangCongQing","description":"🔥3D点云目标检测\u0026语义分割(深度学习)-SOTA方法,代码,论文,数据集等","archived":false,"fork":false,"pushed_at":"2025-01-20T12:00:51.000Z","size":1322,"stargazers_count":510,"open_issues_count":3,"forks_count":78,"subscribers_count":7,"default_branch":"main","last_synced_at":"2025-04-01T07:46:18.635Z","etag":null,"topics":["3d-detection","3d-point-cloud","3d-point-clouds","3d-semantic-segmentation","cpp","dataset","deep-learning","pcl","point-cloud","python3","ros","ros-melodic","sota"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/HuangCongQing.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-01-15T09:38:03.000Z","updated_at":"2025-03-27T01:31:15.000Z","dependencies_parsed_at":"2024-01-16T02:44:28.287Z","dependency_job_id":"9ebeb6ae-e02e-4d57-8696-f4e29b4ffad1","html_url":"https://github.com/HuangCongQing/3D-Point-Clouds","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HuangCongQing%2F3D-Point-Clouds","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HuangCongQing%2F3D-Point-Clouds/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HuangCongQing%2F3D-Point-Clouds/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HuangCongQing%2F3D-Point-Clouds/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/HuangCongQing","download_url":"https://codeload.github.com/HuangCongQing/3D-Point-Clouds/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247801176,"owners_count":20998339,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["3d-detection","3d-point-cloud","3d-point-clouds","3d-semantic-segmentation","cpp","dataset","deep-learning","pcl","point-cloud","python3","ros","ros-melodic","sota"],"created_at":"2024-10-03T10:12:12.430Z","updated_at":"2025-04-08T08:17:49.886Z","avatar_url":"https://github.com/HuangCongQing.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 3D-Point-Clouds\n\n3D点云SOTA方法,代码,论文,数据集(点云目标检测\u0026amp;分割)  \n* 如有疑问，微信：shuangyu_ai\n* 更多自动驾驶相关交流群，欢迎扫码加入：[自动驾驶感知(PCL/ROS+DL)：技术交流群汇总(新版)](https://mp.weixin.qq.com/s?__biz=MzI4OTY1MjA3Mg==\u0026mid=2247486575\u0026idx=1\u0026sn=3145b7a5e9dda45595e1b51aa7e45171\u0026chksm=ec2aa068db5d297efec6ba982d6a73d2170ef09a01130b7f44819b01de46b30f13644347dbf2#rd)\n\n应同学建议，创建了星球 **【自动驾驶感知(PCL/ROS+DL)】** 专注于自动驾驶感知领域，包括传统方法(PCL点云库,ROS)和深度学习（目标检测+语义分割）方法。同时涉及Apollo，Autoware(基于ros2)，BEV感知，三维重建，SLAM(视觉+激光雷达) ，模型压缩（蒸馏+剪枝+量化等），自动驾驶模拟仿真，自动驾驶数据集标注\u0026数据闭环等自动驾驶全栈技术，欢迎扫码二维码加入，一起登顶自动驾驶的高峰！\n\n![image](https://github.com/HuangCongQing/HuangCongQing/assets/20675770/304e0c4d-89d2-4cee-a2a9-3c690611c9d9)\n\n**点云处理方法上主要包括两类方法：**\n\n* 深度学习方法 [`python`]\n  * 目标检测\u0026语义分割\u0026多目标跟踪（MOT）\n  * [【202212done】目标检测最新论文实时更新](https://zhuanlan.zhihu.com/p/591349104)\n  * [【202304done】语义分割最新论文实时更新](https://zhuanlan.zhihu.com/p/591349481)\n* 传统上基于规则的方法 [`c++`]\n  * PCL:https://github.com/HuangCongQing/pcl-learning\n  * ROS:   https://github.com/HuangCongQing/ROS\n  * Apollo笔记：https://github.com/HuangCongQing/apollo_note\n\n@[双愚](https://github.com/HuangCongQing) , 若fork或star请注明来源\n\n## TODO\n\n  - [x] [【202212done】目标检测最新论文实时更新](https://zhuanlan.zhihu.com/p/591349104)\n  - [x] [【202304done】语义分割最新论文实时更新](https://zhuanlan.zhihu.com/p/591349481)\n  - [x] [【202209done】目标检测框架(pcdet+mmdetection3d+det3d+paddle3d)文章撰写](https://zhuanlan.zhihu.com/p/569189196?)\n  - [x] [【202208done】数据集调研总结](https://zhuanlan.zhihu.com/p/551861727)\n  - [x] [【202406done】3D语义分割框架综述(mmdetection3d|OpenPCSeg|Pointcept)](https://zhuanlan.zhihu.com/p/701605684?)\n  - [ ] 数据集详细剖析：kitti\u0026waymo\u0026nuScenes\n  - [ ] Apollo学习https://github.com/HuangCongQing/apollo_note\n\n\n## 目录\n\n#### 0 目标检测框架(pcdet+mmdetection3d+det3d+paddle3d)\n\u003e [【202209done】目标检测框架(pcdet+mmdetection3d+det3d+paddle3d)文章撰写](https://zhuanlan.zhihu.com/p/569189196?)\n\n代码注解笔记：\n1. **pcdet:https://github.com/HuangCongQing/pcdet-note**\n2. **mmdetection3d:https://github.com/HuangCongQing/mmdetection3d-note**\n3. det3d: TODO\n4. paddle3d: TODO\n\n#### 1 paper(code)\n\n* paperswithcode: https://paperswithcode.com/\n\n#### 2 Datasets\n\n**[自动驾驶相关数据集调研总结【附下载地址】(更新ing)](https://zhuanlan.zhihu.com/p/551861727)**\n\n\n数据集基本处理: [数据集标注文件处理](https://github.com/HuangCongQing/Python#%E7%82%B9%E4%BA%91%E7%9B%B8%E5%85%B3%E5%A4%84%E7%90%86)\n\n部分数据下载脚本：https://github.com/HuangCongQing/download_3D_dataset\n\n\n\n\n#### 3 点云可视化\n\n点云可视化笔记和代码：https://github.com/HuangCongQing/Point-Clouds-Visualization\n\n3D点云可视化的库有很多，你的选择可能是：\n\n- pcl 点云可视化 [`c++`]\n- ROS topic可视化  [`c++`] [`python`]\n- open3D [`python`]\n- mayavi[`python`]\n- matplolib [`python`]\n\n\n#### 4 点云数据标注\n数据标注工具总结：https://github.com/HuangCongQing/data-labeling-tools\n\n\n## paper(code)\n\n### 3D_Object_Detection\n- [x]  [**\u003e\u003e\u003e目标检测最新论文实时更新**](https://zhuanlan.zhihu.com/p/591349104)\n\n\n\n* One-stage\n* Two-stage\n\n#### One-stage\n\n\u003e Voxel-Net、SECOND、PointPillars、HVNet、DOPS、Point-GNN、SA-SSD、3D-VID、3DSSD\n\n* Voxel-Net\n* SECOND\n* PointPillars\n* HVNet\n* DOPS\n* Point-GNN\n* SA-SSD\n* 3D-VID\n* 3DSSD\n\n#### Two-stage\n\n\n\u003e F-pointNet、F-ConvNet、Point-RCNN、Part-A^2、PV-RCNN、Fast Point RCNN、TANet\n\n* F-pointNet\n* F-ConvNet\n* Point-RCNN\n* Part-A^2\n* PV-RCNN\n* Fast Point RCNN\n* TANet\n\n### 3D_Semantic_Segmentation\n\n- [x]  [**\u003e\u003e\u003e语义分割最新论文实时更新**](https://zhuanlan.zhihu.com/p/591349481)\n\n\n**PointNet** is proposed to learn per-point features using shared MLPs and global features using symmetrical pooling functions. Based on PointNet, a series of point-based networks have been proposed\n\n\u003ePoint-based Methods: these methods can be roughly divided into pointwise MLP methods, point convolution methods, RNN-based methods, and graph-based methods\n\n#### 1 pointwise MLP methods\n\n\u003e PointNet++，PointSIFT，PointWeb，ShellNet，RandLA-Net\n\n\n\nPointNet++\nPointSIFT\nPointWeb\nShellNet\nRandLA-Net\n\n\n\n\n#### 2 point convolution methods\n\n\u003e PointCNN PCCN A-CNN ConvPoint pointconv KPConv DPC InterpCNN\n\n\n* PointCNN\n* PCCN\n* A-CNN\n* ConvPoint\n* pointconv\n* KPConv\n* DPC\n* InterpCNN\n\n\n#### 3 RNN-based methods\n\u003e G+RCU  RSNet  3P-RNN  DAR-Net\n\n\n* G+RCU  \n* RSNet  \n* 3P-RNN  \n* DAR-Net\n\n\n#### 4 graph-based methods\n\n\u003e DGCNN SPG SSP+SPG PyramNet GACNet PAG HDGCN  HPEIN SPH3D-GCN DPAM\n\n\n* DGCNN\n* SPG\n* SSP+SPG\n* PyramNet\n* GACNet\n* PAG\n* HDGCN\n* HPEIN\n* SPH3D-GCN\n* DPAM\n\n### 3D_Instance Segmentation\n\n\n## Datasets\n\n### 数据集下载\n\n*  **shell脚本下载方式: https://github.com/HuangCongQing/download_3D_dataset**\n\n- [https://hyper.ai/datasets](https://hyper.ai/datasets)\n- [https://www.graviti.cn/open-datasets](https://www.graviti.cn/open-datasets)\n\n\u003e Graviti 收录了 400 多个高质量 CV 类数据集，覆盖无人驾驶、智慧零售、机器人等多种 AI 应用领域。举两个例子：\n\u003e 文章\u003e [https://bbs.cvmart.net/topics/3346](https://bbs.cvmart.net/topics/3346)\n\n- Google数据集搜索：[https://toolbox.google.com/datasetsearch](https://toolbox.google.com/datasetsearch)\n- Datahub，分享高质量数据集平台：[https://datahub.io/](https://datahub.io/)\n- 用于上传和查找数据集的机器学习数据集存储库：[https://www.webdoctx.com/www.mldata.org](https://www.webdoctx.com/www.mldata.org)\n- datafountain收集数据集：[https://www.datafountain.cn/dataSets](https://www.datafountain.cn/dataSets)\n- tinymind收集数据集：[https://www.tinymind.cn/sites#group_22](https://www.tinymind.cn/sites#group_22) 看到的一篇文章,里面有介绍很多数据集的：[世界上最有价值的不是石油而是数据(附数据资源下载链接)](https://mp.weixin.qq.com/s/Ao8SO9j2IPurl45Noy1dVw)\n- [https://www.graviti.cn/open-datasets](https://www.graviti.cn/open-datasets)\n\n## Datasets数据集汇总\n\n[https://github.com/Yochengliu/awesome-point-cloud-analysis#---datasets](https://github.com/Yochengliu/awesome-point-cloud-analysis#---datasets)\n\n- **[**[KITTI](http://www.cvlibs.net/datasets/kitti/)] The KITTI Vision Benchmark Suite. [`det.`]**常用\n- [[ModelNet](http://modelnet.cs.princeton.edu/)] The Princeton ModelNet . [**`cls.`**]\n- [[ShapeNet](https://www.shapenet.org/)] A collaborative dataset between researchers at Princeton, Stanford and TTIC. [**`seg.`**]\n- [[PartNet](https://shapenet.org/download/parts)] The PartNet dataset provides fine grained part annotation of objects in ShapeNetCore. [**`seg.`**]\n- [[PartNet](http://kevinkaixu.net/projects/partnet.html)] PartNet benchmark from Nanjing University and National University of Defense Technology. [**`seg.`**]\n- **[**[**S3DIS**](http://buildingparser.stanford.edu/dataset.html#Download)**] The Stanford Large-Scale 3D Indoor Spaces Dataset. [`seg.`]**常用\n- [[ScanNet](http://www.scan-net.org/)] Richly-annotated 3D Reconstructions of Indoor Scenes. [**`cls.`** **`seg.`**]\n- [[Stanford 3D](https://graphics.stanford.edu/data/3Dscanrep/)] The Stanford 3D Scanning Repository. [**`reg.`**]\n- [[UWA Dataset](http://staffhome.ecm.uwa.edu.au/~00053650/databases.html)] . [**`cls.`** **`seg.`** **`reg.`**]\n- [[Princeton Shape Benchmark](http://shape.cs.princeton.edu/benchmark/)] The Princeton Shape Benchmark.\n- [[SYDNEY URBAN OBJECTS DATASET](http://www.acfr.usyd.edu.au/papers/SydneyUrbanObjectsDataset.shtml)] This dataset contains a variety of common urban road objects scanned with a Velodyne HDL-64E LIDAR, collected in the CBD of Sydney, Australia. There are 631 individual scans of objects across classes of vehicles, pedestrians, signs and trees. [**`cls.`** **`match.`**]\n- [[ASL Datasets Repository(ETH)](https://projects.asl.ethz.ch/datasets/doku.php?id=home)] This site is dedicated to provide datasets for the Robotics community with the aim to facilitate result evaluations and comparisons. [**`cls.`** **`match.`** **`reg.`** **`det`**]\n- [[Large-Scale Point Cloud Classification Benchmark(ETH)](http://www.semantic3d.net/)] This benchmark closes the gap and provides a large labelled 3D point cloud data set of natural scenes with over 4 billion points in total. [**`cls.`**]\n- [[Robotic 3D Scan Repository](http://asrl.utias.utoronto.ca/datasets/3dmap/)] The Canadian Planetary Emulation Terrain 3D Mapping Dataset is a collection of three-dimensional laser scans gathered at two unique planetary analogue rover test facilities in Canada.\n- [[Radish](http://radish.sourceforge.net/)] The Robotics Data Set Repository (Radish for short) provides a collection of standard robotics data sets.\n- [[IQmulus \u0026 TerraMobilita Contest](http://data.ign.fr/benchmarks/UrbanAnalysis/#)] The database contains 3D MLS data from a dense urban environment in Paris (France), composed of 300 million points. The acquisition was made in January 2013. [**`cls.`** **`seg.`** **`det.`**]\n- [[Oakland 3-D Point Cloud Dataset](http://www.cs.cmu.edu/~vmr/datasets/oakland_3d/cvpr09/doc/)] This repository contains labeled 3-D point cloud laser data collected from a moving platform in a urban environment.\n- [[Robotic 3D Scan Repository](http://kos.informatik.uni-osnabrueck.de/3Dscans/)] This repository provides 3D point clouds from robotic experiments，log files of robot runs and standard 3D data sets for the robotics community.\n- [[Ford Campus Vision and Lidar Data Set](http://robots.engin.umich.edu/SoftwareData/Ford)] The dataset is collected by an autonomous ground vehicle testbed, based upon a modified Ford F-250 pickup truck.\n- [[The Stanford Track Collection](https://cs.stanford.edu/people/teichman/stc/)] This dataset contains about 14,000 labeled tracks of objects as observed in natural street scenes by a Velodyne HDL-64E S2 LIDAR.\n- [[PASCAL3D+](http://cvgl.stanford.edu/projects/pascal3d.html)] Beyond PASCAL: A Benchmark for 3D Object Detection in the Wild. [**`pos.`** **`det.`**]\n- [[3D MNIST](https://www.kaggle.com/daavoo/3d-mnist)] The aim of this dataset is to provide a simple way to get started with 3D computer vision problems such as 3D shape recognition. [**`cls.`**]\n- [[WAD](http://wad.ai/2019/challenge.html)] [[ApolloScape](http://apolloscape.auto/tracking.html)] The datasets are provided by Baidu Inc. [**`tra.`** **`seg.`** **`det.`**]\n- [[nuScenes](https://d3u7q4379vrm7e.cloudfront.net/object-detection)] The nuScenes dataset is a large-scale autonomous driving dataset.用过\n- [[PreSIL](https://uwaterloo.ca/waterloo-intelligent-systems-engineering-lab/projects/precise-synthetic-image-and-lidar-presil-dataset-autonomous)] Depth information, semantic segmentation (images), point-wise segmentation (point clouds), ground point labels (point clouds), and detailed annotations for all vehicles and people. [[paper](https://arxiv.org/abs/1905.00160)] [**`det.`** **`aut.`**]\n- [[3D Match](http://3dmatch.cs.princeton.edu/)] Keypoint Matching Benchmark, Geometric Registration Benchmark, RGB-D Reconstruction Datasets. [**`reg.`** **`rec.`** **`oth.`**]\n- [[BLVD](https://github.com/VCCIV/BLVD)] (a) 3D detection, (b) 4D tracking, (c) 5D interactive event recognition and (d) 5D intention prediction. [[ICRA 2019 paper](https://arxiv.org/abs/1903.06405v1)] [**`det.`** **`tra.`** **`aut.`** **`oth.`**]\n- [[PedX](https://arxiv.org/abs/1809.03605)] 3D Pose Estimation of Pedestrians, more than 5,000 pairs of high-resolution (12MP) stereo images and LiDAR data along with providing 2D and 3D labels of pedestrians. [[ICRA 2019 paper](https://arxiv.org/abs/1809.03605)] [**`pos.`** **`aut.`**]\n- [[H3D](https://usa.honda-ri.com/H3D)] Full-surround 3D multi-object detection and tracking dataset. [[ICRA 2019 paper](https://arxiv.org/abs/1903.01568)] [**`det.`** **`tra.`** **`aut.`**]\n- [[Argoverse BY ARGO AI]](https://www.argoverse.org/) Two public datasets (3D Tracking and Motion Forecasting) supported by highly detailed maps to test, experiment, and teach self-driving vehicles how to understand the world around them.[[CVPR 2019 paper](http://openaccess.thecvf.com/content_CVPR_2019/html/Chang_Argoverse_3D_Tracking_and_Forecasting_With_Rich_Maps_CVPR_2019_paper.html)][**`tra.`** **`aut.`**]\n- [[Matterport3D](https://niessner.github.io/Matterport/)] RGB-D: 10,800 panoramic views from 194,400 RGB-D images. Annotations: surface reconstructions, camera poses, and 2D and 3D semantic segmentations. Keypoint matching, view overlap prediction, normal prediction from color, semantic segmentation, and scene classification. [[3DV 2017 paper](https://arxiv.org/abs/1709.06158)] [[code](https://github.com/niessner/Matterport)] [[blog](https://matterport.com/blog/2017/09/20/announcing-matterport3d-research-dataset/)]\n- [[SynthCity](https://arxiv.org/abs/1907.04758)] SynthCity is a 367.9M point synthetic full colour Mobile Laser Scanning point cloud. Nine categories. [**`seg.`** **`aut.`**]\n- [[Lyft Level 5](https://level5.lyft.com/dataset/?source=post_page)] Include high quality, human-labelled 3D bounding boxes of traffic agents, an underlying HD spatial semantic map. [**`det.`** **`seg.`** **`aut.`**]\n- **[**[**SemanticKITTI**](http://semantic-kitti.org/)**] Sequential Semantic Segmentation, 28 classes, for autonomous driving. All sequences of KITTI odometry labeled. [**[**ICCV 2019 paper**](https://arxiv.org/abs/1904.01416)**] [`seg.` `oth.` `aut.`]**常用\n- [[NPM3D](http://npm3d.fr/paris-lille-3d)] The Paris-Lille-3D has been produced by a Mobile Laser System (MLS) in two different cities in France (Paris and Lille). [**`seg.`**]\n- [[The Waymo Open Dataset](https://waymo.com/open/)] The Waymo Open Dataset is comprised of high resolution sensor data collected by Waymo self-driving cars in a wide variety of conditions. [**`det.`**]\n- [[A*3D: An Autonomous Driving Dataset in Challeging Environments](https://github.com/I2RDL2/ASTAR-3D)] A*3D: An Autonomous Driving Dataset in Challeging Environments. [**`det.`**]\n- [[PointDA-10 Dataset](https://github.com/canqin001/PointDAN)] Domain Adaptation for point clouds.\n- [[Oxford Robotcar](https://robotcar-dataset.robots.ox.ac.uk/)] The dataset captures many different combinations of weather, traffic and pedestrians. [**`cls.`** **`det.`** **`rec.`**]\n\n### 常用分割数据集\n\n- **[**[**S3DIS**](http://buildingparser.stanford.edu/dataset.html#Download)**] The Stanford Large-Scale 3D Indoor Spaces Dataset. [`seg.`] [`常用`]\n- **[**[**SemanticKITTI**](http://semantic-kitti.org/)**] Sequential Semantic Segmentation, 28 classes, for autonomous driving. All sequences of KITTI odometry labeled. [**[**ICCV 2019 paper**](https://arxiv.org/abs/1904.01416)**] [`seg.` `oth.` `aut.`] [`常用`]\n- **Semantic3d**\n\n### 常用分类数据集\n\ntodo\n\n### 常用目标检测数据集\n\n- **[**[KITTI](http://www.cvlibs.net/datasets/kitti/)] The KITTI Vision Benchmark Suite. [`det.`]**常用\n- [[nuScenes](https://d3u7q4379vrm7e.cloudfront.net/object-detection)] The nuScenes dataset is a large-scale autonomous driving dataset.用过\n- [[The Waymo Open Dataset](https://waymo.com/open/)] The Waymo Open Dataset is comprised of high resolution sensor data collected by Waymo self-driving cars in a wide variety of conditions. [**`det.`**]\n\n## References\n\n* https://github.com/timzhang642/3D-Machine-Learning\n* https://github.com/victorphd/autonomous-vahicles-learning-resource\n* https://github.com/Yochengliu/awesome-point-cloud-analysis\n* https://github.com/NUAAXQ/awesome-point-cloud-analysis-2021\n* https://github.com/QingyongHu/SoTA-Point-Cloud\n* https://arxiv.org/abs/1912.12033 : Deep Learning for 3D Point Clouds: A Survey\n* https://github.com/zhulf0804/3D-PointCloud\n\n## License\n\nCopyright (c) [双愚](https://github.com/HuangCongQing). All rights reserved.\n\nLicensed under the [MIT](./LICENSE) License.\n\n---\n\n\n微信公众号：**【双愚】**（huang_chongqing） 聊科研技术,谈人生思考,欢迎关注~\n\n![image](https://user-images.githubusercontent.com/20675770/169835565-08fc9a49-573e-478a-84fc-d9b7c5fa27ff.png)\n\n**往期推荐：**\n1. [本文不提供职业建议，却能助你一生](https://mp.weixin.qq.com/s/rBR62qoAEeT56gGYTA0law)\n2. [聊聊我们大学生面试](https://mp.weixin.qq.com/s?__biz=MzI4OTY1MjA3Mg==\u0026mid=2247484016\u0026idx=1\u0026sn=08bc46266e00572e46f3e5d9ffb7c612\u0026chksm=ec2aae77db5d276150cde1cb1dc6a53e03eba024adfbd1b22a048a7320c2b6872fb9dfef32aa\u0026scene=178\u0026cur_album_id=2253272068899471368#rd)\n3. [清华大学刘知远：好的研究方法从哪来](https://mp.weixin.qq.com/s?__biz=MzI4OTY1MjA3Mg==\u0026mid=2247486340\u0026idx=1\u0026sn=6c5f69bb37d91a343b1a1e7f6929ddae\u0026chksm=ec2aa783db5d2e95ba4c472471267721cafafbe10c298a6d5fae9fed295f455a72f783872249\u0026scene=178\u0026cur_album_id=1855544495514140673#rd)\n\n本人创建星球 **【自动驾驶感知(PCL/ROS+DL)】** 专注于自动驾驶感知领域，包括传统方法(PCL点云库,ROS)和深度学习（目标检测+语义分割）方法。同时涉及Apollo，Autoware(基于ros2)，BEV感知，三维重建，SLAM(视觉+激光雷达) ，模型压缩（蒸馏+剪枝+量化等），自动驾驶模拟仿真，自动驾驶数据集标注\u0026数据闭环等自动驾驶全栈技术，欢迎扫码二维码加入，一起登顶自动驾驶的高峰！\n![image](https://github.com/HuangCongQing/HuangCongQing/assets/20675770/304e0c4d-89d2-4cee-a2a9-3c690611c9d9)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhuangcongqing%2F3d-point-clouds","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhuangcongqing%2F3d-point-clouds","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhuangcongqing%2F3d-point-clouds/lists"}