Non-Gibbsian Markov random field models for contextual labelling of structured scenes

In this paper we propose a non-Gibbsian Markov random field to model the spatial and topological relationships between objects in s tructured scenes. The field is formulated in terms of conditional probabilities le arned from a set of training images. A locally consistent labelling of new scenes is achieved by relaxing the Markov random field directly using these condit ional probabilities. We evaluate our model on a varied collection of several hundred handsegmented images of buildings.