Copyright Vladimir Kolmogorov vnk@ist.ac.at 2013

POTTS, version 1.1.
Implements the algorithm in 

	I. Gridchyn, V. Kolmogorov.
	"Potts model, parametric maxflow and k-submodular functions". 
	In IEEE International Conference on Computer Vision (ICCV), Sydney, Australia, December 2013.

Note: for experiments in the paper we used the FastPD software for the second stage:

	N. Komodakis and G. Tziritas.
	"Approximate labeling via graph cuts based on linear programming."
	PAMI, 29(8):1436-1453, 2007.

	N. Komodakis, G. Tziritas, and N. Paragios.
	"Performance vs computational efficiency for optimizing single and dynamic MRFs: Setting the state of the art with primal-dual strategies."
	CVIU, 112(1):14-29, 2008

FastPD is not included in this software; instead, a more naive version of the alpha-expansion algorithm is implemented.


Based on the Boykov-Kolmogorov maxflow algorithm, with flow and search tree recycling (as in [Kohli,Torr ICCV'05]).



For usage see Potts.h. Create a new instance of Potts, specify unary terms, add edges and then call Potts::Solve().
