This platform provide a real-time monocular SLAM method that computes the camera trajectory and a sparse 3D reconstruction by leveraging point (ORB) and line (LSD) features. We provide examples to run the system on the ICL NUIM dataset.
If you are interested in more structure information about indoor SLAM, we command you to check our PlanarSLAM that proposes ManhattanWorld/VanishingDirection and other modules.
PL-SLAM is released under a GPLv3 license. For a closed-source version of Structure-SLAM(PL) for commercial purposes, please contact me yanyan.li at tum.de
We have tested the library in Ubuntu 16.04, but it should be easy to compile in other platforms. A powerful computer (e.g. i7) will ensure real-time performance and provide more stable and accurate results.
We use the new thread and chrono functionalities of C++11.
We use Pangolin for visualization and user interface. Dowload and install instructions can be found at: https://github.com/stevenlovegrove/Pangolin.
We use OpenCV to manipulate images and features. Dowload and install instructions can be found at: http://opencv.org. Required at least 2.4.3. Tested with OpenCV 3.4.0.
Required by g2o (see below). Download and install instructions can be found at: http://eigen.tuxfamily.org. Required at least 3.1.0.
We use modified versions of the DBoW2 library to perform place recognition and g2o library to perform non-linear optimizations. Both modified libraries (which are BSD) are included in the Thirdparty folder.
We provide a script build.sh to build the Thirdparty libraries and Structure-SLAM. Please make sure you have installed all required dependencies (see section 2). Execute:
cd Structure-SLAM
chmod +x build.sh
./build.sh
- Download ICL NUIM dataset and uncompress it to PATH_TO_SEQUENCE_FOLDER
- Execute the following command.
./Examples/Structure-SLAM Vocabulary/ORBvoc.txt Examples/ICL.yaml PATH_TO_SEQUENCE_FOLDER
This platform is a part of Structure-SLAM, please cite it if you use the repo in an academic work.
@inproceedings{Li2020SSLAM,
author = {Li, Yanyan and Brasch, Nikolas and Wang, Yida and Navab, Nassir and Tombari, Federico},
title = {Structure-SLAM: Low-Drift Monocular SLAM in Indoor Environments},
year = {2020},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
}
@article{li2020rgb,
title={RGB-D SLAM with Structural Regularities},
author={Li, Yanyan and Yunus, Raza and Brasch, Nikolas and Navab, Nassir and Tombari, Federico},
journal={IEEE International Conference on Robotics and Automation (ICRA)},
year={2021}
}
We thank Raul Mur-Artal for his impressive work, ORB-SLAM2, which is a completed feature-based SLAM system.