Description:

OpenGM is a free C++ template library, a command line tool and a set of MATLAB functions for optimization in higher order graphical models. Graphical models of any order and structure can be built either in C++ or in MATLAB, using simple and intuitive commands. These models can be stored in HDF5 files and optimized using state-of-the-art algorithms and the OpenGM command line optimizer. All library functions can also be called directly from C++ code. OpenGM realizes the Inference Algorithm Interface (IAI), a concept that makes it easy for programmers to use their own algorithms and factor classes with OpenGM.

• Factor Graph Models (Kschischang et al. 2001)

  • Graphs of any order and structure, from regular grid graphs to irregular graphical models with higher order factors.
  • Flexible number of labels (different variables can have differently many labels).

• Optimization Algorithms

  • Loopy Belief Propagation (Pearl 1988, Yedidia et al. 2000) with message damping (Wainwright 2008), including Min-Sum and Max-Product message passing.
  • Tree-reweighted Belief Propagation (TRBP) (Wainwright et al. 2005) with message damping.
  • A-star branch-and-bound search (Bergtholdt et al. 2009).
  • Sub-Gradient Descent (Dual Decomposition) coming soon! (Kappes et al. 2010).
    • Automated decomposition of arbitrary factor graphs
  • Iterated conditional modes (ICM) (Besag 1986).
  • Lazy Flipper (Andres et al. 2010). Binary variables only.
  • Graph Cut (Boycov et al. 2001). Push-Relabel (Goldberg and Tarjan 1986), Edmonds-Karp (Edmonds and Karp 1972), Kolmogorov (Boykov and Kolmogorov 2004]). Binary variables, 2nd order models and submodular functions only.

• Command Line Optimizer

  • Built-in protocol mode for runtime and convergence analyses.

• MATLAB Interface

  • Build your graphical models conveniently in MATLAB.
  • HDF5 Import/Export

• High Performance Computing

  • Optimization of graphical models that consist of 10^7 factors and more.
  • Optimized class templates for binary variables (contributed by Thorben Kroeger).

• Extendibility

  • Add and contribute your own optimization algorithms and factor classes.

Changes to previous version:

Initial Announcement on mloss.org.

BibTeX Entry: Download

Corresponding Paper BibTeX Entry: Download

Supported Operating Systems: Platform Independent

Data Formats: Hdf

Tags: Bayesian Networks, Factor Graphs, Graphical Models, Markov Random Fields, Belief Propagation, Discrete Optimization, Higher Order Cliques, Higher Order Factors, Inference

Archive: download here

Other available revisons

Version	Changelog	Date
1.0 -- Optimization Library for Higher Order Graphical Models	Initial Announcement on mloss.org.	November 12, 2010, 17:00:50
1.0	Initial Announcement on mloss.org.	November 12, 2010, 16:58:40
1.0 beta	Initial Announcement on mloss.org.	November 12, 2010, 15:30:55

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.