Fusion Moves for Markov Random Field Optimization

DMCA

Fusion Moves for Markov Random Field Optimization

Cached

Download Links

by Victor Lempitsky , Carsten Rother , Stefan Roth , Andrew Blake

Citations:

68 - 5 self

BibTeX

@MISC{Lempitsky_fusionmoves,
    author = {Victor Lempitsky and Carsten Rother and Stefan Roth and Andrew Blake},
    title = { Fusion Moves for Markov Random Field Optimization},
    year = {}
}

OpenURL

Abstract

The efficient application of graph cuts to Markov Random Fields (MRFs) with multiple discrete or continuous labels remains an open question. In this paper, we demonstrate one possible way of achieving this by using graph cuts to combine pairs of suboptimal labelings or solutions. We call this combination process the fusion move. By employing recently developed graph cut based algorithms (so-called QPBO-graph cut), the fusion move can efficiently combine two proposal labelings in a theoretically sound way, which is in practice often globally optimal. We demonstrate that fusion moves generalize many previous graph cut approaches, which allows them to be used as building block within a broader variety of optimization schemes than were considered before. In particular, we propose new optimization schemes for computer vision MRFs with applications to image restoration, stereo, and optical flow, among others. Within these schemes the fusion moves are used 1) for the parallelization of MRF optimization into several threads; 2) for fast MRF optimization by combining cheapto-compute solutions; and 3) for the optimization of highly non-convex continuous-labeled MRFs with 2D labels. Our final example is a non-vision MRF concerned with cartographic label placement, where fusion moves can be used to improve the performance of a standard inference method (loopy belief propagation).

Keyphrases

fusion move graph cut markov random field optimization markov random field continuous label building block multiple discrete computer vision mrfs open question loopy belief propagation suboptimal labelings combination process so-called qpbo-graph cut non-vision mrf possible way final example mrf optimization non-convex continuous-labeled mrfs many previous graph cut approach optimization scheme several thread standard inference method new optimization scheme fast mrf optimization cartographic label placement proposal labelings efficient application optical flow cheapto-compute solution sound way image restoration

DMCA