Awesome

StereoGlue

More examples and fitting problems are coming soon!

Installation

Clone the repository and its submodules:

git clone https://github.com/danini/stereoglue.git

Make sure that you have the necessary libraries installed:

sudo apt-get install libopencv-dev libopencv-contrib-dev libarpack++2-dev libarpack2-dev libsuperlu-dev libeigen3-dev libboost-all-dev pybind11

To run affine feature detection and matching with the built-in tools, install:

pip install kornia
pip install kornia-moons

git clone https://github.com/cvg/LightGlue.git && cd LightGlue
python -m pip install -e .

Install StereoGlue by running

pip install .

Requirements

• Eigen 3.0 or higher
• CMake 2.8.12 or higher
• OpenCV 3.0 or higher
• A modern compiler with C++17 support

Evaluation

Jupyter Notebook examples

The example for essential matrix fitting with gravity-based solver is available at: notebook.

The example for fundamental matrix fitting with monodepth-based solver is available at: notebook.

The weights for the affine correspondence estimation can be downloaded from link.

The model for feature orientation and scale estimation can be downloaded from link.

Acknowledgements

When using the algorithm, please cite:

@inproceedings{StereoGlue2024,
	author = {Barath, Daniel and Mishkin, Dmytro and Cavalli, Luca and Sarlin, Paul-Edouard and Hruby, Petr and Pollefeys, Marc},
	title = {{StereoGlue}: Robust Estimation with Single-Point Solvers},
	booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
	year = {2024},
}

If you use it for fundamental matrix estimation, please cite:

@inproceedings{Hruby2024,
	author = {Hruby, Petr and Pollefeys, Marc and Barath, Daniel},
	title = {Semicalibrated Relative Pose from an Affine Correspondence and Monodepth},
	booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
	year = {2024},
}

If you use it with MAGSAC++ scoring, please cite:

@inproceedings{barath2020magsac++,
  title={MAGSAC++, a fast, reliable and accurate robust estimator},
  author={Barath, Daniel and Noskova, Jana and Ivashechkin, Maksym and Matas, Jiri},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages={1304--1312},
  year={2020}
}