PCL（Point Cloud Library）点云库 个人开发环境：Ubuntu18.04

<!-- * @Description: * @Author: HCQ * @Company(School): UCAS * @Date: 2020-10-04 18:17:00 * @LastEditors: Please set LastEditors * @LastEditTime: 2023-08-13 12:09:52 --> # pcl   [](https://github.com/murufeng/awesome_lightweight_networks) [](https://github.com/murufeng/awesome_lightweight_networks)  PCL（Point Cloud Library）点云库 **个人开发环境：Ubuntu18.04** * 如有疑问，微信：shuangyu_ai * **Plus: _可承接点云处理相关项目，欢迎私聊_** * **PCL微信交流群二维码每周都更新一次，请关注公众号【双愚】后台回复PCL加群** * 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==&mid=2247486575&idx=1&sn=3145b7a5e9dda45595e1b51aa7e45171&chksm=ec2aa068db5d297efec6ba982d6a73d2170ef09a01130b7f44819b01de46b30f13644347dbf2#rd) * [**【本仓库测试pcd数据】**](./data) **墙裂建议先看下：[PCL(Point Cloud Library)学习指南&资料推荐](https://zhuanlan.zhihu.com/p/268524083)** **<font color='red'>PCL学习入门指南&代码实践(最新版)入门视频： </font> https://www.bilibili.com/video/BV1HS4y1y7AB** **代码对应系列笔记：[PCL(Point Cloud Library)学习记录（2024）](https://www.yuque.com/huangzhongqing/pcl)** <img src="https://github.com/HuangCongQing/pcl-learning/assets/20675770/f55c8b20-d7ec-43df-9f53-4271a0dcab12" alt="微信交流群二维码" width="50%;" height="50%" /> **相关项目实战:** * 3D-MOT(多目标检测和追踪): [https://github.com/HuangCongQing/3D-LIDAR-Multi-Object-Tracking/tree/kitti](https://github.com/HuangCongQing/3D-LIDAR-Multi-Object-Tracking/tree/kitti) * 需要学习ROS：https://github.com/HuangCongQing/ROS @[双愚](https://github.com/HuangCongQing/pcl-learning) , 若fork或star请注明来源 > * 点云数据的处理可以采用获得广泛应用的Point Cloud Library (点云库，PCL库)。 > * PCL库是一个最初发布于2013年的开源C++库。它实现了大量点云相关的通用算法和高效的数据管理。 > * 支持多种操作系统平台，可在Windows、Linux、Android、Mac OS X、部分嵌入式实时系统上运行。如果说OpenCV是2D信息获取与处理的技术结晶，那么PCL在3D信息获取与处理上，就与OpenCV具有同等地位 > * PCL是BSD授权方式，可以免费进行商业和学术应用。 * 英文官网：https://pcl.readthedocs.io/projects/tutorials/en/latest/# * https://pointclouds.org/ * GitHub：https://github.com/PointCloudLibrary/pcl * 学习基于pcl1.9.1：https://github.com/PointCloudLibrary/pcl/tree/pcl-1.9.1 **Tips:** * ubuntu下使用PCL，需要写**CMakeLists.txt**文件，然后编译才可以生成可执行文件. * 可执行文件在build文件夹下，所以运行可执行文件时，后面添加参数的pcd文件，应放在build文件夹下才能获取到。**（注意文件路径）** * `make -j ` (-j 自动多线程， -j4 四线程) ## 目录contents > ***建议必学** * [**【本仓库测试pcd数据】**](./data) * [00base](00base) ##### step1 * [01common](01common ) ##### step2 * [* 02kdtree k维tree](02kdtree) [[doc](https://www.yuque.com/huangzhongqing/pcl/uffamg#w9i1y)] * [* 03octree 八叉树](03octree) [[doc](https://www.yuque.com/huangzhongqing/pcl/habl9h)] * [* 04search](04search): [[doc](https://www.yuque.com/huangzhongqing/pcl/qs4wx2)] * [05sample consensus 抽样一致性模块](05sampleconsensus抽样一致性模块) [[doc](https://www.yuque.com/huangzhongqing/pcl/ivtxgx)] * [06range-images深度图像](06range-images深度图像) [[doc](https://www.yuque.com/huangzhongqing/pcl/hxeyrz)] ##### step3（must） * [* 08 io 输入输出](08IO输入输出) [[doc](https://www.yuque.com/huangzhongqing/pcl/mt2yo5)] * [* 09 filters 滤波](09filters滤波) [[doc](https://www.yuque.com/huangzhongqing/pcl/ai96k5)] * [* 10 features 特征](10features特征) [[doc](https://www.yuque.com/huangzhongqing/pcl/kf9kmf)] ##### step4（根据个人需要） * [11 surface表面 ](11surface表面) [[doc](https://www.yuque.com/huangzhongqing/pcl/yfrd0w)] * [12 segmentation分割](12segmentation分割) [[doc](https://www.yuque.com/huangzhongqing/pcl/kg7wvi)] * [13 recognition识别](13recognition识别) [[doc](https://www.yuque.com/huangzhongqing/pcl/hpgc39)] * [14 registration配准](14registration配准) [[doc](https://www.yuque.com/huangzhongqing/pcl/zg7alz)] * [* 15 visualization可视化](15visualization可视化) [[doc](https://www.yuque.com/huangzhongqing/pcl/rmexll)] * [16 keypoints关键点](16keypoints关键点) [[doc](https://www.yuque.com/huangzhongqing/pcl/twi0mt)] * [07tracking跟踪](07tracking跟踪/tracking.md) [[doc](https://www.yuque.com/huangzhongqing/pcl/em72xa)] ## 编译过程 ```shell mkdir build cd build cmake .. // 对上一级进行编译 make // 生成可执行文件 ./executedemo // 运行可执行文件 ``` ## 实战项目 不理解的地方,欢迎提issue: https://github.com/HuangCongQing/pcl-learning/issues * 3D-MOT(多目标检测和追踪) * https://github.com/HuangCongQing/3D-LIDAR-Multi-Object-Tracking/tree/kitti * 3D点云目标检测&语义分割-SOTA方法,代码,论文,数据集等 * https://github.com/HuangCongQing/3D-Point-Clouds ## 相关链接 * 公众号：点云PCL * https://github.com/Yochengliu/awesome-point-cloud-analysis * https://github.com/QingyongHu/SoTA-Point-Cloud * https://github.com/PointCloudLibrary/pcl * 参考书籍：点云库PCL学习教程，朱德海，北京航空航天大学出版社 * Plus：ROS学习-https://github.com/HuangCongQing/ROS **入门资料：** - **<font color='red'>PCL学习入门指南&代码实践(最新版)入门视频： </font> https://www.bilibili.com/video/BV1HS4y1y7AB** - **视频**：[bilibili-PCL点云库官网教程](https://space.bilibili.com/504859351/channel/detail?cid=130387) - **点云库PCL学习教程书籍每章总结：**[https://github.com/MNewBie/PCL-Notes](https://github.com/MNewBie/PCL-Notes) - 百度网盘资料： 链接：[https://pan.baidu.com/s/1ziq8s_kj5QpM8eXO_d6RJg](https://pan.baidu.com/s/1ziq8s_kj5QpM8eXO_d6RJg)<br />提取码：g6ny<br /> **代码实践资料：** - 官方各模块示例(和对应的对象函数对照着看)【英文】：[https://pcl.readthedocs.io/projects/tutorials/en/latest/#](https://pcl.readthedocs.io/projects/tutorials/en/latest/#) - 官方各模块对应的对象和函数【英文】： - [https://pointclouds.org/documentation/modules.html](https://pointclouds.org/documentation/modules.html) - [https://pointclouds.org/](https://pointclouds.org/) 点击网站中的12宫图，没一格对应一个模块的对象函数详解 - [黑马机器人系列文档：PCL-3D点云](http://robot.czxy.com/docs/pcl/)：[http://robot.czxy.com/docs/pcl/](http://robot.czxy.com/docs/pcl/) - [CSDN博主系列文章PCL学习(64篇)](https://www.cnblogs.com/li-yao7758258/category/954066.html)：[https://www.cnblogs.com/li-yao7758258/category/954066.html](https://www.cnblogs.com/li-yao7758258/category/954066.html) ## Citation If you find this project useful in your research, please consider cite: ``` @misc{pcl-learning2020, title={A Complete Study Guide on How to Learn PCL (Point Cloud Library).}, author={Chongqing, Huang}, howpublished = {\url{https://github.com/HuangCongQing/pcl-learning}}, year={2020} } ``` ### 欢迎交流 后续会根据逐步完善，欢迎大家提出宝贵意见，也欢迎大家提issue pr�