patchwork++复现————————地面点云分割复现

如有问题可以联系我v：15130883468

代码地址：大但你kdurl-kaist/patchwork-plusplus：Patchwork++：用于 3D LiDAR 扫描的快速而强大的地面分割方法。@ IROS'22 (github.com)

我使用的虚拟机复现的ubuntu 20.04，当然18.04也是可以的。

首先运行下边的命令

# Install prerequisite packages including Open3D
#克隆代码
$ git clone https://github.com/url-kaist/patchwork-plusplus
$ cd patchwork-plusplus
$ bash scripts/install_open3d.bash

但是

bash scripts/install_open3d.bash是自动下载指令往往会运行失败

这个是时候就需要手动下载

# To install Eigen and numpy
#注意numpy>=1.23.0
$ sudo apt-get install libeigen3-dev
$pip install numpy==1.23

# To install Open3D Python packages
$ pip install open3d

# To install Open3D C++ packages

$ git clone https://github.com/isl-org/Open3D
$ cd Open3D
$ util/install_deps_ubuntu.sh # Only needed for Ubuntu
$ mkdir build && cd build
$ cmake ..


#make指令会出现较多错误推荐使用sudo make  或者make -j4(or -j8)这个时候出现错误可以根据错误手动下载，可以参考我的另一篇文章很简单。

$make  # If it fails, try several times or try 'sudo make'



$ sudo make install

构建

在源代码文件下打开

 patchwork-plusplus

$ mkdir build && cd build
$ cmake ..
$ make

运行：

# Run patchwork++ and visualize ground points(green) and nonground points(red)
$ ./build/examples/cpp/demo_visualize

# Run patchwork++ with sequential point cloud inputs 
$ ./build/examples/cpp/demo_sequential

# Run patchwork++ with your point cloud file, example here
$ ./build/examples/cpp/demo_visualize ./data/000000.bin # specify file path

效果展示

这个现在只能使用c++构建

python构建总是显示 

ERROR: Could not build wheels for pypatchworkpp, which is required to install pyproject.toml-based projects
 

如果只是显示地面地面点云或者非地面点云可以更改

patchwork-plusplus/examples/cpp/demo_visualize.cpp文件的

  visualizer.AddGeometry(geo_ground);            //地面部分的点云显示
  visualizer.AddGeometry(geo_nonground);         //非地面部分点云显示



//-------------------注释完成后重新运行在patchwork-plusplus/build路径下打开命令窗口运行make指令

根据需要注释，注释完成后重新运行在patchwork-plusplus/build路径下打开命令窗口运行make指令。最后回到patchwork-plusplus路径。执行命令

cd patchwork-plusplus/build
make 

#返回上一级目录执行代码
cd ..            
./build/examples/cpp/demo_visualize ./data/000001.bin 

显示结果如下