https://github.com/masadcv/numpymaxflow

Numpy-based implementation of Max-flow/Min-cut (graphcut) for 2D/3D data
https://github.com/masadcv/numpymaxflow

numpy python segmentation

Last synced: 1 day ago
JSON representation

Numpy-based implementation of Max-flow/Min-cut (graphcut) for 2D/3D data

• Host: GitHub
• URL: https://github.com/masadcv/numpymaxflow
• Owner: masadcv
• License: bsd-3-clause
• Created: 2022-04-19T18:53:23.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-10-30T21:16:25.000Z (16 days ago)
• Last Synced: 2024-11-14T14:11:11.632Z (1 day ago)
• Topics: numpy, python, segmentation
• Language: C++
• Homepage: https://pypi.org/project/numpymaxflow/
• Size: 2.69 MB
• Stars: 2
• Watchers: 1
• Forks: 4
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# numpymaxflow: Max-flow/Min-cut in numpy for 2D images and 3D volumes

[![License](https://img.shields.io/badge/License-BSD_3--Clause-blue.svg)](https://opensource.org/licenses/BSD-3-Clause)



[![CI Build](https://github.com/masadcv/numpymaxflow/actions/workflows/build.yml/badge.svg)](https://github.com/masadcv/numpymaxflow/actions/workflows/build.yml)

[![PyPI version](https://badge.fury.io/py/numpymaxflow.svg)](https://badge.fury.io/py/numpymaxflow)





Numpy-based implementation of Max-flow/Min-cut based on the following paper:

- Boykov, Yuri, and Vladimir Kolmogorov. "An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision." IEEE transactions on pattern analysis and machine intelligence 26.9 (2004): 1124-1137.

If you want same functionality in PyTorch, then consider [PyTorch-based implementation](https://github.com/masadcv/torchmaxflow)

## Citation

If you use this code in your research, then please consider citing:

 **Asad, Muhammad, Lucas Fidon, and Tom Vercauteren. ["ECONet: Efficient Convolutional Online Likelihood Network for Scribble-based Interactive Segmentation."](https://openreview.net/pdf?id=9xtE2AgD_Cc) Medical Imaging with Deep Learning (MIDL), 2022.**

## Installation instructions

`pip install numpymaxflow`

or 

```

# Clone and install from github repo

$ git clone https://github.com/masadcv/numpymaxflow

$ cd numpymaxflow

$ pip install -r requirements.txt

$ python setup.py install

```

## Example outputs

Maxflow2d

![./figures/numpymaxflow_maxflow2d.png](https://raw.githubusercontent.com/masadcv/numpymaxflow/main/figures/numpymaxflow_maxflow2d.png)

Interactive maxflow2d

![./figures/numpymaxflow_intmaxflow2d.png](https://raw.githubusercontent.com/masadcv/numpymaxflow/main/figures/numpymaxflow_intmaxflow2d.png)

![figures/figure_numpymaxflow.png](https://raw.githubusercontent.com/masadcv/numpymaxflow/main/figures/figure_numpymaxflow.png)

## Example usage

The following demonstrates a simple example showing numpymaxflow usage:

```python

image = np.asarray(Image.open('data/image2d.png').convert('L'), np.float32)

image = np.expand_dims(image, axis=0)

prob = np.asarray(Image.open('data/image2d_prob.png'), np.float32)

lamda = 20.0

sigma = 10.0

post_proc_label = numpymaxflow.maxflow(image, prob, lamda, sigma)

```

For more usage examples see: 

**2D and 3D maxflow and interactive maxflow examples**: [`demo_maxflow.py`](https://raw.githubusercontent.com/masadcv/numpymaxflow/main/demo_maxflow.py) 

 

## References

- [OpenCV's Graphcut implementation](https://github.com/opencv/opencv/blob/4.x/modules/imgproc/include/opencv2/imgproc/detail/gcgraph.hpp)

- [SimpleCRF's maxflow implementation](https://github.com/HiLab-git/SimpleCRF)

- [torchmaxflow's implementation](https://github.com/masadcv/torchmaxflow)

This repository depends on the code for [maxflow from latest version of OpenCV](https://github.com/opencv/opencv/blob/4.x/modules/imgproc/include/opencv2/imgproc/detail/gcgraph.hpp), which has been included.