Name
Title | ||
|---|---|---|
What Is Optimized in Convex Relaxations for Multilabel Problems: Connecting Discrete and Continuously Inspired MAP Inference |
Abstract | ||
|---|---|---|
In this work, we present a unified view on Markov random fields (MRFs) and recently proposed continuous tight convex relaxations for multilabel assignment in the image plane. These relaxations are far less biased toward the grid geometry than Markov random fields on grids. It turns out that the continuous methods are nonlinear extensions of the well-established local polytope MRF relaxation. In view of this result, a better understanding of these tight convex relaxations in the discrete setting is obtained. Further, a wider range of optimization methods is now applicable to find a minimizer of the tight formulation. We propose two methods to improve the efficiency of minimization. One uses a weaker, but more efficient continuously inspired approach as initialization and gradually refines the energy where it is necessary. The other one reformulates the dual energy enabling smooth approximations to be used for efficient optimization. We demonstrate the utility of our proposed minimization schemes in numerical experiments. Finally, we generalize the underlying energy formulation from isotropic metric smoothness costs to arbitrary nonmetric and orientation dependent smoothness terms. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/TPAMI.2013.105 | Pattern Analysis and Machine Intelligence, IEEE Transactions |
Keywords | Field | DocType |
Markov processes,approximation theory,image processing,inference mechanisms,maximum likelihood estimation,optimisation,smoothing methods,Markov random fields,grid geometry,image plane,inspired MAP inference,local polytope MRF relaxation,multilabel assignment,multilabel problems,optimization methods,smooth approximations,tight convex relaxations,Markov random fields,approximate inference,continuous labeling problems,convex relaxation | Mathematical optimization,Random field,Markov process,Computer science,Markov chain,Approximate inference,Polytope,Variable-order Markov model,Initialization,Smoothness | Journal |
Volume | Issue | ISSN |
36 | 1 | 0162-8828 |
Citations | PageRank | References |
12 | 0.64 | 27 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Christopher Zach | 1 | 1457 | 84.01 |
| Christian Hane | 2 | 281 | 17.03 |
| Marc Pollefeys | 3 | 7671 | 475.90 |