Implemented fast grow cut algorithm based on "An Effective Interactive Medical Image Segmentation Method Using Fast GrowCut" (see link), generalized for multi-class n-dimensional data.
Run setup file (ONLY needed for Cython version, NOT required for the other Python implementation)
python setup.py build
sudo python setup.py install
For basic GrowCut package usase, see (link).
Fast Growcut using shortest path
Implemented in growcut_cpu
Run fastgc function with image data img, initial seed labels seeds, both img and seeds can be n-dimensional data.
fastgc(img, seeds, newSeg = True, labCrt=None, distCrt=None, labPre=None, distPre=None, verbose = True)
Run pytest directly
Test 1: test_growcut.py