MIT-SPARK/KISS-Matcher — Gitpedia

<div align="center"> <h1>KISS-Matcher</h1> <a href="https://github.com/MIT-SPARK/KISS-Matcher"><img src="https://img.shields.io/badge/-C++-blue?logo=cplusplus" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher"><img src="https://img.shields.io/badge/Python-3670A0?logo=python&logoColor=ffdd54" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher"><img src="https://img.shields.io/badge/ROS2-Humble-blue" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher"><img src="https://img.shields.io/badge/Ubuntu-E95420?logo=ubuntu&logoColor=white" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher"><img src="https://img.shields.io/badge/macOS-000000?logo=apple&logoColor=white" /></a> <a href="https://arxiv.org/abs/2409.15615"><img src="https://img.shields.io/badge/arXiv-b33737?logo=arXiv" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher/actions/workflows/ci-cpp.yml"><img src="https://github.com/MIT-SPARK/KISS-Matcher/actions/workflows/ci-cpp.yml/badge.svg?branch=main" alt="C++ Core Build" /></a> <a href="https://github.com/MIT-SPARK/KISS-Matcher/actions/workflows/ci-python.yml"><img src="https://github.com/MIT-SPARK/KISS-Matcher/actions/workflows/ci-python.yml/badge.svg?branch=main" alt="Python Build" /></a> <img src="https://github.com/user-attachments/assets/763bafef-c11a-4412-a9f7-f138fc12ff9f" alt="KISS Matcher" width="95%"/> Keep it simple, make it scalable. (May 19, 2026) Python package is now available on PyPI <a href="https://pypi.org/project/kiss-matcher/"><img src="https://readme-typing-svg.demolab.com?font=Fira+Code&weight=500&size=17&duration=2200&pause=900&color=E5E7EB&background=111827&center=true&vCenter=true&width=460&height=48&lines=%E2%9D%AF+pip+install+kiss-matcher;%E2%9C%93+ready+for+Python" alt="pip install kiss-matcher"/></a> </div>

:package: Installation

All installations are set up automatically in an out-of-the-box manner.

Run the command below to install basic dependency packages:

commandline
git clone https://github.com/MIT-SPARK/KISS-Matcher.git
cd KISS-Matcher
make deps

make deps auto-detects the host OS:

Linux (Ubuntu/Debian): installs the toolchain via apt-get.
macOS (Apple Silicon / Intel): installs the toolchain via Homebrew.
AppleClang does not ship with OpenMP, so we pull in llvm + libomp
from Homebrew and the Makefile automatically routes the C++ build through
$(brew --prefix llvm)/bin/clang++. If you have not installed Homebrew yet,
install it first from https://brew.sh.

After that, follow the C++ or Python installation instructions below.

C++

Run the command below. That's it.

commandline
git clone https://github.com/MIT-SPARK/KISS-Matcher.git
cd KISS-Matcher
make cppinstall

[!WARNING]
If you encounter the following error:
CMake Error: Not a file: ${your_directory}/KISS-Matcher/cpp/kiss_matcher/build/_deps/robin-build/cmake_install.cmake
CMake Error: Error processing file: ${your_directory}/KISS-Matcher/cpp/kiss_matcher/build/_deps/robin-build/cmake_install.cmake
make: *** [Makefile:42: cppinstall] Error 1
it means that the ROBIN package is already installed in your environment.
In that case, simply run the following command:
make cppinstall_matcher_only

<details> <summary>Q. How doest it work?</a></summary>

The cppinstall command is encapsulated in the Makefile. and cppinstall calls deps to automatically install the dependencies.
In addition, KISS-Matcher requires ROBIN.
But it's also automatically linked via robin.cmake

After installation, kiss_matcher and robin are placed in the installation directory using the following commands in the Makefile, respectively:
@$(SUDO) cmake --install cpp/kiss_matcher/build @$(SUDO) cmake --install cpp/kiss_matcher/build/_deps/robin-build

</details> <details> <summary>Q. Building from a non-Makefile flow on macOS?</summary>

If you invoke cmake directly on macOS instead of going through make cppinstall, point CMake at the Homebrew LLVM toolchain so OpenMP is available:

bash
LLVM_PREFIX=$(brew --prefix llvm)
cmake -S cpp/kiss_matcher -B build \
      -DCMAKE_BUILD_TYPE=Release \
      -DCMAKE_C_COMPILER="$LLVM_PREFIX/bin/clang" \
      -DCMAKE_CXX_COMPILER="$LLVM_PREFIX/bin/clang++" \
      -DUSE_SYSTEM_ROBIN=OFF
cmake --build build -j$(sysctl -n hw.ncpu)

USE_SYSTEM_ROBIN=OFF forces a fresh ROBIN via FetchContent and avoids
picking up an older / incompatible system install. On Linux this flag is not
required, but it is harmless.

</details>

Example codes

We provide plentiful scalable registration examples. Please visit our cpp/example directory and follow the instructions.

How To Use in Other Packages in C++?

See CMakeLists.txt in cpp/examples.

Python

Install from PyPI (Linux x86_64 manylinux_2_28+, macOS 14+ arm64; sdist build elsewhere):

bash
pip install kiss-matcher           # core only (numpy)
pip install kiss-matcher[viz]      # adds viser for the visualization demo

Or install editably from a source checkout:

bash
pip install --upgrade pip setuptools wheel scikit-build-core ninja cmake build
pip install -e python/

Quickstart

After installing, run the synthetic-data smoke test (no external data, no viser):

bash
python python/examples/quickstart.py

It generates 5k random points, applies a known rigid transform, and prints the
recovered rotation/translation errors. Expect sub-degree rotation error and
sub-centimetre translation error.

For real point clouds with 3D visualization (requires [viz] extra):

bash
python python/examples/run_kiss_matcher.py \
    --src_path /path/to/src.pcd \
    --tgt_path /path/to/tgt.pcd \
    --resolution 0.3

We also provide out-of-the-box python registration examples. Go to python directory and follow the instructions.

Citation

This study is the culmination of my continuous research since my graduate school years.
If you use this library for any academic work, please cite our original paper, and refer to other papers that are highly relevant to KISS-Matcher.

<details> <summary>See bibtex lists</a></summary>

bibtex
 @inproceedings{lim2025icra-KISSMatcher,
   title={{KISS-Matcher: Fast and Robust Point Cloud Registration Revisited}},
   author={Lim, Hyungtae and Kim, Daebeom and Shin, Gunhee and Shi, Jingnan and Vizzo, Ignacio and Myung, Hyun and Park, Jaesik and Carlone, Luca},
   booktitle={Proc. IEEE Int. Conf. Robot. Automat.},
   year={2025},
   codeurl   = {https://github.com/MIT-SPARK/KISS-Matcher},
   note   = {Accepted. To appear}
 }

bibtex
 @article{Lim24ijrr-Quatropp,
   title={{Quatro++: R}obust global registration exploiting ground segmentation for loop closing in {LiDAR SLAM}},
   author={Lim, Hyungtae and Kim, Beomsoo and Kim, Daebeom and Mason Lee, Eungchang and Myung, Hyun},
   journal={Int. J. Robot. Res.},
   pages={685--715},
   year={2024},
   doi={10.1177/02783649231207654}
 }

bibtex
@inproceedings{Lim22icra-Quatro,
    title={A single correspondence is enough: Robust global registration to avoid degeneracy in urban environments},
    author={Lim, Hyungtae and Yeon, Suyong and Ryu, Soohyun and Lee, Yonghan and Kim, Youngji and Yun, Jaeseong and Jung, Euigon and Lee, Donghwan and Myung, Hyun},
    booktitle={Proc. IEEE Int. Conf. Robot. Automat.},
    pages={8010--8017},
    year={2022}
}

bibtex
@InProceedings{Shi21icra-robin,
    title={{ROBIN:} a Graph-Theoretic Approach to Reject Outliers in Robust Estimation using Invariants},
    author={J. Shi and H. Yang and L. Carlone},
    booktitle={Proc. IEEE Int. Conf. Robot. Automat.},
    note = {arXiv preprint: 2011.03659, \linkToPdf{https://arxiv.org/pdf/2011.03659.pdf}},
    pdf="https://arxiv.org/pdf/2011.03659.pdf",
    year={2021}
}

bibtex
@article{Yang20tro-teaser,
    title={{TEASER: Fast and Certifiable Point Cloud Registration}},
    author={H. Yang and J. Shi and L. Carlone},
    journal={IEEE Trans. Robot.},
    volume = 37,
    number = 2,
    pages = {314--333},
    note = {extended arXiv version 2001.07715 \linkToPdf{https://arxiv.org/pdf/2001.07715.pdf}},
    pdf={https://arxiv.org/pdf/2001.07715.pdf},
    Year = {2020}
}

bibtex
@inproceedings{Zhou16eccv-FGR,
    title={Fast global registration},
    fullauthor={Zhou, Qian-Yi and Park, Jaesik and Koltun, Vladlen},
    author={Q.Y. Zhou and J. Park and V. Koltun},
    booktitle={Proc. Eur. Conf. Comput. Vis.},
    pages={766--782},
    year={2016}
}

</details>

Contributing

Like KISS-ICP,
we envision KISS-Matcher as a community-driven project, we love to see how the project is growing thanks to the contributions from the community. We would love to see your face in the list below, just open a Pull Request!

Acknowledgements

This project was supported by the NRF grant funded by the government of South Korea, Ministry of Science and ICT (MSIT) (No. RS-2024-00461409).

We sincerely express our gratitude to Prof. Cyrill Stachniss’ group at the University of Bonn for their generosity in allowing us to use the term KISS.
In particular, I (Hyungtae Lim) personally deeply appreciate Tizziano Guadagnino, Benedikt Mersch, Louis Wiesmann, and Jens Behley for their kind support and collaboration during my time at the University of Bonn.
Without them, KISS-Matcher would not exist today. A special thanks goes to my co-author, Nacho, for taking the time to thoroughly teach me modern C++ and modern CMake.
We would also like to express our gratitude to Kenji Koide for his exceptional effort in open-sourcing the wonderful open source, small_gicp, which plays a crucial role in enhancing the efficiency of our pipeline.
His dedication to the research community continues to inspire and enable advancements in the field.