Entropy Controlled Gauss-Markov Random Measure Field Models for Early Vision

We present a computationally efficient segmentationrestoration method, based on a probabilistic formulation, for the joint estimation of the label map (segmentation) and the parameters of the feature generator models (restoration). Our algorithm computes an estimation of the posterior marginal probability distributions of the label field based on a Gauss Markov Random Measure Field model. Our proposal introduces a explicit entropy control for the estimated posterior marginals, therefore improving the parameter estimation step. If the model parameters are given, our algorithm computes the posterior marginals as the global minimizer of a quadratic, linearly constrained energy function; therefore, one can compute very efficiently the optimal (Maximizer of the Posterior Marginals or MPM) estimator for multi–class segmentation problems. Moreover, a good estimation of the posterior marginals allows one to compute estimators different from the MPM for restoration problems, denoising and optical flow computation. Experiments demonstrate better performance over other state of the art segmentation approaches.