The Sim Cut Algorithm

The Sim Cut is a important advance in image segmentation. The use of the maximum-flow theory to obtain image segmentation, first reported in 1998, has been found to be highly effective and the application to medical imaging has been widespread. The Sim Cut provides a strikingly new method for computing this type of image segmentation that exploits the Frisch Theorem; a theorem that shows that maximum-flow problems can be solved in an approximate manner by analog electrical circuits. The algorithm is compatible with massively parallel computing and has applications in image segmentation and stereo reconstruction.

Learn more:

US Patent 8929636

Technical Report: Precision of the multiway cut using Vector Electrical Flow

Technical Report: Acceleration of the Graph Cut with High Performance Computing

Technical Report: A Vector Electrical Network for Multi-way Graph Partition

Sim Cut on AMAZON LAMBDA

PETSc demonstration of the Sim Cut

Python demonstration of the Sim Cut
Copyright © 2012 Virtual Scalpel, Inc., All rights reserved