Hierarchy-Based-Segmentation

Python implementation of our paper "Assessing hierarchies by their consistent segmentations".

Authors : Zeev Gutman, Ritvik Vij, Laurent Najman, Michael Lindenbaum

Usage

A Jupyter notebook src/Interactive.ipynb contains the user-friendly code for running the experiments given in the paper.

All helper scripts and implementation of the auxiliary algorithms are given in the src directory.

The data is placed in the data directory. The HED gradient images are in HED and SLIC superpixels are stored in SLIC.

Using your own data

You can easily load your custom images in the interactive notebook.

To generate your own superpixels and to generate your own gradient images, you can use the scripts in Helper_Scripts

cd Helper_Scripts
bash run_all_slic.sh
bash run_hed_all.sh

Requirements

numpy=1.16.4
higra=0.5.3
numba=0.51.2
scipy=1.5.2
matplotlib=3.3.2
opencv-contrib-python=4.1.2.30

Cite

Please cite our paper if you use the code or ideas in your own work

@unpublished{gutman:hal-03633805,
  TITLE = {{Assessing hierarchies by their consistent segmentations}},
  AUTHOR = {Gutman, Zeev and Vij, Ritvik and Najman, Laurent and Lindenbaum, Michael},
  URL = {https://hal.archives-ouvertes.fr/hal-03633805},
  NOTE = {working paper or preprint},
  YEAR = {2022},
  MONTH = Apr,
  PDF = {https://hal.archives-ouvertes.fr/hal-03633805/file/hierarchy_based_segmentation.pdf},
  HAL_ID = {hal-03633805},
  HAL_VERSION = {v1},
}

Contact

For any communication related to the code or the paper, feel free to contact me at [email protected].