A MRF Shape Prior for Facade Parsing with Occlusions

Mateusz Koziński (1, 2, 3) , Raghudeep Gadde (3, 4, 1, 2) , Sergey Zagoruyko (1, 2, 3) , Guillaume Obozinski (1, 3, 2) , Renaud Marlet (2, 1, 3)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 ENPC - École des Ponts ParisTech
3 imagine [Marne-la-Vallée]
4 CVN - Centre de vision numérique

We present a new shape prior formalism for the segmentation of rectified facade images. It combines the simplicity of split grammars with unprecedented expressive power: the capability of encoding simultaneous alignment in two dimensions, facade occlusions and irregular boundaries between facade elements. We formulate the task of finding the most likely image segmentation conforming to a prior of the proposed form as a MAP-MRF problem over a 4-connected pixel grid, and propose an efficient optimization algorithm for solving it. Our method simultaneously segments the visible and occluding objects, and recovers the structure of the occluded facade. We demonstrate state-of-the-art results on a number of facade segmentation datasets.

Keywords: Shape; Image segmentation; Grammar; Encoding; Production; Semantics; Layout

Domains: Computer Vision and Pattern Recognition [cs.CV]

Mateusz Koziński, Raghudeep Gadde, Sergey Zagoruyko, Guillaume Obozinski, Renaud Marlet. A MRF Shape Prior for Facade Parsing with Occlusions. Conference on Computer Vision and Pattern Recognition (CVPR 2015), Jun 2015, Boston, Massachusetts, United States. pp.2820-2828, ⟨10.1109/CVPR.2015.7298899⟩. ⟨hal-01232598⟩