- I am happy to announce that Graph-Cut RANSAC had been included in OpenCV. You can check the documentation at link.
The Graph-Cut RANSAC algorithm proposed in paper: Daniel Barath and Jiri Matas; Graph-Cut RANSAC, Conference on Computer Vision and Pattern Recognition, 2018. It is available at https://openaccess.thecvf.com/content_cvpr_2018/papers/Barath_Graph-Cut_RANSAC_CVPR_2018_paper.pdf
The journal paper with updated spatial coherence term and additional experiments is available at https://ieeexplore.ieee.org/document/9399280
The method is explained in the Latest developments in RANSAC presentation from CVPR tutorial RANSAC in 2020.
Experiments on homography, fundamental matrix, essential matrix, and 6D pose estimation are shown in the corresponding presentation from the tutorial RANSAC in 2020.
There are pre-compiled wheels for Windows and Linux for Python 3.8 - 3.11 that you can install with:
pip install pygcransac
Thanks to @akaszynski for his contributions.
To build and install C++ only GraphCutRANSAC
, clone or download this repository and then build the project by CMAKE.
$ git clone https://github.com/danini/graph-cut-ransac
$ cd build
$ cmake ..
$ make
python3 ./setup.py install
or
pip3 install -e .
To build the sample project showing examples of fundamental matrix, homography and essential matrix fitting, set variable CREATE_SAMPLE_PROJECT = ON
when creating the project in CMAKE.
Then
$ cd build
$ ./SampleProject
- Eigen 3.0 or higher
- CMake 2.8.12 or higher
- OpenCV 3.0 or higher
- A modern compiler with C++17 support
import pygcransac
h1, w1 = img1.shape
h2, w2 = img2.shape
H, mask = pygcransac.findHomography(src_pts, dst_pts, h1, w1, h2, w2, 3.0)
F, mask = pygcransac.findFundamentalMatrix(src_pts, dst_pts, h1, w1, h2, w2, 3.0)
The example for homography fitting is available at: notebook.
The example for fundamental matrix fitting is available at: notebook.
The example for essential matrix fitting is available at: notebook.
The example for essential matrix fitting with planar motion assumption is available at: notebook.
The example for essential matrix fitting with known gravity is available at: notebook.
The example for essential matrix fitting is available at: notebook.
The example for 6D pose fitting is available at: notebook.
An example comparing different samplers on fundamental matrix fitting: notebook.
The example for homography fitting to SIFT correspondences is available at: notebook.
The example for essential matrix fitting to SIFT correspondences is available at: notebook.
The example for fundamental matrix fitting to SIFT correspondences is available at: notebook.
The example for homography fitting to affine correspondences is available at: notebook.
The example for essential matrix fitting to affine correspondences is available at: notebook.
The example for fundamental matrix fitting to affine correspondences is available at: notebook.
- Python 3
- CMake 2.8.12 or higher
- OpenCV 3.4
- A modern compiler with C++11 support
When using the algorithm, please cite
@inproceedings{GCRansac2018,
author = {Barath, Daniel and Matas, Jiri},
title = {Graph-cut {RANSAC}},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2018},
}
If you use it together with Progressive NAPSAC sampling or DEGENSAC, please cite
@inproceedings{PNAPSAC2020,
author = {Barath, Daniel and Noskova, Jana and Ivashechkin, Maksym and Matas, Jiri},
title = {{MAGSAC}++, a Fast, Reliable and Accurate Robust Estimator},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}
@inproceedings{Degensac2005,
author = {Chum, Ondrej and Werner, Tomas and Matas, Jiri},
title = {Two-View Geometry Estimation Unaffected by a Dominant Plane},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2005},
}
If you use with Space Partitioning turned on, please cite
@inproceedings{spaceransac2022,
author = {Barath, Daniel and Valasek, Gabor},
title = {Space-Partitioning {RANSAC}},
journal={European Conference on Computer Vision},
year = {2022}
}
The Python wrapper part is based on the great Benjamin Jack python_cpp_example
.