fast_gicp

Note: New faster library is released

We released small_gicp that is twice as fast as fast_gicp and with minimum dependencies and clean interfaces.

fast_gicp

This package is a collection of GICP-based fast point cloud registration algorithms. It constains a multi-threaded GICP as well as multi-thread and GPU implementations of our voxelized GICP (VGICP) algorithm. All the implemented algorithms have the PCL registration interface so that they can be used as an inplace replacement for GICP in PCL.

• FastGICP: multi-threaded GICP algorithm (~40FPS)
• FastGICPSingleThread: GICP algorithm optimized for single-threading (~15FPS)
• FastVGICP: multi-threaded and voxelized GICP algorithm (~70FPS)
• FastVGICPCuda: CUDA-accelerated voxelized GICP algorithm (~120FPS)
• NDTCuda: CUDA-accelerated D2D NDT algorithm (~500FPS)

on melodic & noetic

Installation

Dependencies

• PCL
• Eigen
• OpenMP
• CUDA (optional)
• Sophus
• nvbio

We have tested this package on Ubuntu 18.04/20.04 and CUDA 11.1.

On macOS when using brew, you might have to set up your depenencies like this

cmake .. "-DCMAKE_PREFIX_PATH=$(brew --prefix libomp)[;other-custom-prefixes]" -DQt5_DIR=$(brew --prefix qt@5)lib/cmake/Qt5

CUDA

To enable the CUDA-powered implementations, set BUILD_VGICP_CUDA cmake option to ON.

ROS

cd ~/catkin_ws/src
git clone https://github.com/SMRT-AIST/fast_gicp --recursive
cd .. && catkin_make -DCMAKE_BUILD_TYPE=Release
# enable cuda-based implementations
# cd .. && catkin_make -DCMAKE_BUILD_TYPE=Release -DBUILD_VGICP_CUDA=ON

Non-ROS

git clone https://github.com/SMRT-AIST/fast_gicp --recursive
mkdir fast_gicp/build && cd fast_gicp/build
cmake .. -DCMAKE_BUILD_TYPE=Release
# enable cuda-based implementations
# cmake .. -DCMAKE_BUILD_TYPE=Release -DBUILD_VGICP_CUDA=ON
make -j8

Python bindings

cd fast_gicp
python3 setup.py install --user

Note: If you are on a catkin-enabled environment and the installation doesn't work well, comment out find_package(catkin) in CMakeLists.txt and run the above installation command again.

…

Benchmark

CPU:Core i9-9900K GPU:GeForce RTX2080Ti

roscd fast_gicp/data
rosrun fast_gicp gicp_align 251370668.pcd 251371071.pcd

…

See src/align.cpp for the detailed usage.

Test on KITTI

C++

# Perform frame-by-frame registration
rosrun fast_gicp gicp_kitti /your/kitti/path/sequences/00/velodyne

Python

cd fast_gicp/src
python3 kitti.py /your/kitti/path/sequences/00/velodyne

Note

In some environments, setting a fewer number of threads rather than the (default) maximum number of threads may result in faster processing (see https://github.com/SMRT-AIST/fast_gicp/issues/145#issuecomment-1890885373).

Related packages

• ndt_omp
• fast_gicp

Papers

• Kenji Koide, Masashi Yokozuka, Shuji Oishi, and Atsuhiko Banno, Voxelized GICP for fast and accurate 3D point cloud registration, ICRA2021 [link]

Contact

Kenji Koide, [email protected]

Human-Centered Mobility Research Center, National Institute of Advanced Industrial Science and Technology, Japan [URL]