MapClosures

  •   Install   •   Paper   •   Contact Us

Effectively Detecting Loop Closures using Point Cloud Density Maps.

---

Install

Dependencies

• Essentials

  sudo apt-get install --no-install-recommends -y build-essential cmake pybind11-dev python3-dev python3-pip

• Optionally Built
  In this case you have two options:

  • Option 1: You can install them by the package manager in your operative system, e.g. in Ubuntu 22.04:

    sudo apt-get install libeigen3-dev libopencv-dev libtbb-dev

    this will of course make the build of MapClosures much faster.
  • Option 2: Let the build system handle them:

    cmake -B build -S cpp -DUSE_SYSTEM_TBB=OFF -DUSE_SYSTEM_OPENCV=OFF -DUSE_SYSTEM_EIGEN3=OFF

    this will be slower in terms of build time, but will enable you to have a different version of this libraries installed in your system without interfering with the build of MapClosures.

Once the dependencies are installed, the C++ library can be build by using standard cmake commands:

cmake -B build -S cpp
cmake --build build -j8

Python Package

We provide a python wrapper for MapClosures which can be easily installed by simply running:

make

Usage

The following command will provide details about how to use our pipeline:

map_closure_pipeline --help

Citation

If you use this library for any academic work, please cite our original paper.

@inproceedings{gupta2024icra,
    author     = {S. Gupta and T. Guadagnino and B. Mersch and I. Vizzo and C. Stachniss},
    title      = {{Effectively Detecting Loop Closures using Point Cloud Density Maps}},
    booktitle  = {IEEE International Conference on Robotics and Automation (ICRA)},
    year       = {2024},
    codeurl    = {https://github.com/PRBonn/MapClosures},
}

Paper Results

As we decided to continue the development of MapClosures beyond the scope of the ICRA paper, we created a git tag so that researchers can consistently reproduce the results of the publication. To checkout at this tag, you can run the following:

git checkout ICRA2024

Our development aims to push the performances of MapClosures above the original results of the paper.

Acknowledgement

This repository is heavily inspired by, and also depends on KISS-ICP

Contributors