Discrete Optimization of the Multiphase Piecewise Constant Mumford-Shah Functional

The Mumford-Shah model has been one of the most powerful models in image segmentation and denoising. The optimization of the multiphase Mumford-Shah energy functional has been performed using level sets methods that optimize the Mumford-Shah energy by evolving the level sets via the gradient descent. These methods are very slow and prone to getting stuck in local optima due to the use of the gradient descent. After the reformulation of the bimodal Mumford-Shah functional on a graph, several groups investigated the hierarchical extension of the graph representation to multi class. These approaches, though more effective than level sets, provide approximate solutions and can diverge away from the optimal solution. In this paper, we present a discrete optimization for the multiphase Mumford Shah functional that directly minimizes the multiphase functional without recursive bisection on the labels. Our approach handles the nonsubmodularity of the multiphase energy function and provide a global optimum if prior information is provided.