International Journal of Computer Vision

, Volume 115, Issue 2, pp 155–184

A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems

  • Jörg H. Kappes
  • Bjoern Andres
  • Fred A. Hamprecht
  • Christoph Schnörr
  • Sebastian Nowozin
  • Dhruv Batra
  • Sungwoong Kim
  • Bernhard X. Kausler
  • Thorben Kröger
  • Jan Lellmann
  • Nikos Komodakis
  • Bogdan Savchynskyy
  • Carsten Rother
Article

DOI: 10.1007/s11263-015-0809-x

Cite this article as:
Kappes, J.H., Andres, B., Hamprecht, F.A. et al. Int J Comput Vis (2015) 115: 155. doi:10.1007/s11263-015-0809-x

Abstract

Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov random fields. This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically, the models today often include higher order interactions, flexible connectivity structures, large label-spaces of different cardinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of more than 27 state-of-the-art optimization techniques on a corpus of 2453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.

Keywords

Discrete graphical models Combinatorial optimization  Benchmark 

Supplementary material

11263_2015_809_MOESM1_ESM.pdf (4 mb)
Supplementary material 1 (pdf 4053 KB)

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Jörg H. Kappes
    • 1
  • Bjoern Andres
    • 2
  • Fred A. Hamprecht
    • 1
  • Christoph Schnörr
    • 1
  • Sebastian Nowozin
    • 3
  • Dhruv Batra
    • 4
  • Sungwoong Kim
    • 5
  • Bernhard X. Kausler
    • 1
  • Thorben Kröger
    • 1
  • Jan Lellmann
    • 6
  • Nikos Komodakis
    • 7
  • Bogdan Savchynskyy
    • 8
  • Carsten Rother
    • 8
  1. 1.Heidelberg UniversityHeidelbergGermany
  2. 2.Combinatorial Image Analysis, Max Planck Institute for InformaticsSaarbrückenGermany
  3. 3.Machine Learning and Perception, Microsoft ResearchCambridgeUnited Kingdom
  4. 4.Virginia Tech, 302 Whittemore HallBlacksburgUSA
  5. 5.Qualcomm Research Korea, 15th FL., POBA Gangnam TowerSeoulRepublic of Korea
  6. 6.DAMTP, University of CambridgeCambridgeUnited Kingdom
  7. 7.Universite Paris-Est, Ecole des Ponts ParisTech, Cité DescartesChamps-sur-MarneFrance
  8. 8.Dresden University of TechnologyDresdenGermany

Personalised recommendations