Felzenszwalb-supervoxel-segmentation

Extension of the Felzenszwalb scikit-image superpixel algorithm to 3D supervoxels.

Compile felzenszwalb_3d_cy.pyx with Cython by running: python setup.py build_ext --inplace

This implementation was developed as part of the ADNet project:

@article{hansen2022anomaly,
  title={Anomaly Detection-Inspired Few-Shot Medical Image Segmentation Through Self-Supervision With Supervoxels},
  author={Hansen, Stine and Gautam, Srishti and Jenssen, Robert and Kampffmeyer, Michael},
  journal={Medical Image Analysis},
  pages={102385},
  year={2022},
  publisher={Elsevier}
}

Please consider citing it if you find our implementation useful.