Crowd-SLAM is a visual SLAM system that is robust in crowded scenarios.
Demonstration video: https://www.youtube.com/watch?v=LeS8MEVaR2E
Paper: https://link.springer.com/article/10.1007/s10846-021-01414-1
Crowd-SLAM is released under a GPLv3 License.
If you use Crowd-SLAM in an academic work, please cite:
@article{soaresJINT2021,
title={Crowd-{SLAM}: Visual {SLAM} Towards Crowded Environments using Object Detection},
author={Soares, J. C. V., Gattass, M. and Meggiolaro, M. A.},
journal={Journal of Intelligent & Robotic Systems},
volume={102},
number={50},
doi = {https://doi.org/10.1007/s10846-021-01414-1},
year={2021}
}
- Install ORB-SLAM2 prerequisites: C++11 or C++0x Compiler, Pangolin, OpenCV and Eigen3 (https://github.com/raulmur/ORB_SLAM2).
- Clone the repository:
git clone https://github.com/virgolinosoares/Crowd-SLAM
- Execute:
cd Crowd-SLAM
chmod +x build.sh
./build.sh
We have tested the library in Ubuntu 18.04, with OpenCV 3.4.
-
Download a sequence from https://vision.in.tum.de/data/datasets/rgbd-dataset/download and uncompress it.
-
Associate RGB images and depth images using the python script associate.py:
python associate.py PATH_TO_SEQUENCE/rgb.txt PATH_TO_SEQUENCE/depth.txt > associations.txt
-
Execute the following command. Change
TUMX.yaml
to TUM1.yaml,TUM2.yaml or TUM3.yaml for freiburg1, freiburg2 and freiburg3 sequences respectively. ChangePATH_TO_SEQUENCE_FOLDER
to the uncompressed sequence folder. ChangeASSOCIATIONS_FILE
to the path to the corresponding associations file../rgbd Vocabulary/ORBvoc.txt Examples/RGB-D/TUMX.yaml PATH_TO_SEQUENCE_FOLDER ASSOCIATIONS_FILE
Our code builds on ORB-SLAM2.