Skip to Main content Skip to Navigation
Conference papers

Generalized Fast Approximate Energy Minimization via Graph Cuts: Alpha-Expansion Beta-Shrink Moves

Mark Schmidt 1, 2 Karteek Alahari 2, 3
1 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR8548
3 WILLOW - Models of visual object recognition and scene understanding
DI-ENS - Département d'informatique de l'École normale supérieure, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR8548
Abstract : We present alpha-expansion beta-shrink moves, a simple generalization of the widely-used alpha-beta swap and alpha-expansion algorithms for approximate energy minimization. We show that in a certain sense, these moves dominate both alpha-beta-swap and alpha-expansion moves, but unlike previous generalizations the new moves require no additional assumptions and are still solvable in polynomial-time. We show promising experimental results with the new moves, which we believe could be used in any context where alpha-expansions are currently employed.
Document type :
Conference papers
Complete list of metadata

Cited literature [18 references]  Display  Hide  Download

https://hal.inria.fr/inria-00617524
Contributor : Mark Schmidt <>
Submitted on : Monday, August 29, 2011 - 2:16:35 PM
Last modification on : Tuesday, May 4, 2021 - 2:06:02 PM
Long-term archiving on: : Tuesday, November 13, 2012 - 9:35:49 AM

Files

sample2e.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00617524, version 1
  • ARXIV : 1108.5710

Collections

Citation

Mark Schmidt, Karteek Alahari. Generalized Fast Approximate Energy Minimization via Graph Cuts: Alpha-Expansion Beta-Shrink Moves. UAI 2011 - 27th Conference on Uncertainty in Artificial Intelligence, Jul 2011, Barcelona, Spain. ⟨inria-00617524⟩

Share

Metrics

Record views

610

Files downloads

2396