Tightening MRF Relaxations with Planar Subproblems

02/14/2012
by   Julian Yarkony, et al.
0

We describe a new technique for computing lower-bounds on the minimum energy configuration of a planar Markov Random Field (MRF). Our method successively adds large numbers of constraints and enforces consistency over binary projections of the original problem state space. These constraints are represented in terms of subproblems in a dual-decomposition framework that is optimized using subgradient techniques. The complete set of constraints we consider enforces cycle consistency over the original graph. In practice we find that the method converges quickly on most problems with the addition of a few subproblems and outperforms existing methods for some interesting classes of hard potentials.

READ FULL TEXT
research
04/06/2011

Planar Cycle Covering Graphs

We describe a new variational lower-bound on the minimum energy configur...
research
12/23/2018

End-to-end Learning for Graph Decomposition

We propose a novel end-to-end trainable framework for the graph decompos...
research
03/05/2011

Submodular Decomposition Framework for Inference in Associative Markov Networks with Global Constraints

In the paper we address the problem of finding the most probable state o...
research
09/23/2017

A Simple Algorithm for Computing a Cycle Separator

We present a linear time algorithm for computing a cycle separator in a ...
research
10/16/2012

Efficiently Searching for Frustrated Cycles in MAP Inference

Dual decomposition provides a tractable framework for designing algorith...
research
08/02/2012

Fast Planar Correlation Clustering for Image Segmentation

We describe a new optimization scheme for finding high-quality correlati...

Please sign up or login with your details

Forgot password? Click here to reset