• Stars
    star
    125
  • Rank 286,335 (Top 6 %)
  • Language
    Python
  • License
    MIT License
  • Created about 8 years ago
  • Updated over 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Python implementation of KNN Matting, CVPR 2012 / TPAMI 2013 http://dingzeyu.li/projects/knn/

knn-matting

Python port of source code for KNN Matting, CVPR 2012 / TPAMI 2013 http://dingzeyu.li/projects/knn/

Requirements

  • python 3.5 (Though it should run on 2.7)
  • scipy
  • numpy
  • matplotlib
  • sklearn

Running the demo

  • '''python knn_matting.py'''
  • mylambda (λ) is a constant controlling the users confidence in the constraints
  • image size not larger than 640*480 reccomended for speed and memory reasons.

Results

Original image

Trimap image

Result image

More Information

The sparse matrix solve step may be greatly sped up by using the scikit-umfpack wrapper of UMFPACK sparse direct solver, https://github.com/scikit-umfpack/scikit-umfpack
For more information see the orginal paper and it's website here http://dingzeyu.li/projects/knn/ The original matlab code is here https://github.com/dingzeyuli/knn-matting

Disclaimer

The code is free for academic/research purpose. Use at your own risk and we are not responsible for any loss resulting from this code. Feel free to submit pull request for bug fixes.

Contact

Marco Forte ([email protected])

Original authors:
Qifeng Chen ([email protected]) and Dingzeyu Li ([email protected])