Inference by Learning: Speeding-up Graphical Model Optimization via a Coarse-to-Fine Cascade of Pruning Classifiers

Bruno Conejo (1, 2) , Nikos Komodakis (3, 1, 4) , Sebastien Leprince (2) , Jean-Philippe Avouac (2)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 Division of Geological and Planetary Sciences [Pasadena]
3 imagine [Marne-la-Vallée]
4 ENPC - École des Ponts ParisTech

We propose a general and versatile framework that significantly speeds-up graph-ical model optimization while maintaining an excellent solution accuracy. The proposed approach, refereed as Inference by Learning or in short as IbyL, relies on a multi-scale pruning scheme that progressively reduces the solution space by use of a coarse-to-fine cascade of learnt classifiers. We thoroughly experiment with classic computer vision related MRF problems, where our novel framework constantly yields a significant time speed-up (with respect to the most efficient inference methods) and obtains a more accurate solution than directly optimizing the MRF. We make our code available on-line [4].

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

Bruno Conejo, Nikos Komodakis, Sebastien Leprince, Jean-Philippe Avouac. Inference by Learning: Speeding-up Graphical Model Optimization via a Coarse-to-Fine Cascade of Pruning Classifiers. NIPS, Dec 2014, Montreal, Canada. ⟨hal-01081590⟩