Semi-global matching: A principled derivation in terms of message passing

Amnon Drory, Carsten Haubold, Shai Avidan, Fred A. Hamprecht

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Semi-global matching, originally introduced in the context of dense stereo, is a very successful heuristic to minimize the energy of a pairwise multi-label Markov Random Field defined on a grid. We offer the first principled explanation of this empirically successful algorithm, and clarify its exact relation to belief propagation and tree-reweighted message passing. One outcome of this new connection is an uncertainty measure for the MAP label of a variable in a Markov Random Field.

Original languageEnglish
Title of host publicationPattern Recognition - 36th German Conference, GCPR 2014, Proceedings
EditorsJoachim Hornegger, Xiaoyi Jiang, Reinhard Koch
Number of pages11
ISBN (Electronic)9783319117515
StatePublished - 2014
Event36th German Conference on Pattern Recognition, GCPR 2014 - Münster, Germany
Duration: 2 Sep 20145 Sep 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)


Conference36th German Conference on Pattern Recognition, GCPR 2014

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Semi-global matching: A principled derivation in terms of message passing'. Together they form a unique fingerprint.

Cite this