Tum Technische Universität München

We propose a general class of label configuration priors for continuous multi-label optimization problems. In contrast to MRF-based approaches, the proposed framework unifies label configuration energies such as minimum description length priors, co-occurrence priors and hierarchical label cost priors. Moreover, it does not require any preprocessing in terms of super-pixel estimation. All problems are solved using efficient primal-dual algorithms which scale better with the number of labels than the alphaexpansion method commonly used in the MRF setting. Experimental results confirm that label configuration priors lead to drastic improvements in segmentation. In particular, the hierarchical prior allows to jointly compute a semantic segmentation and a scene classification.