Factor Graph Models (Kschischang et al. 2001)
    Graphs of any order and structure, from second order grid graphs to irregular higher-order models
    Arbitrary (commutative and associative) operations, including sum, product, conjunction and disjunction
    Flexible number of labels (different variables can have differently many labels)
    Function sharing across factors
    Function type abstraction. Different (built-in and custom) encodings can be used alongside each other
Functions
    Explicit function (multi-dimensional table)
    Sparse function (sparse multi-dimensional table)
    Potts functions (different types, including higher-order)
    Truncated absolute difference
    Truncated squared difference
    Views that treat one graphical model as a function within another graphical model
Algorithms
    Loopy Belief Propagation (Pearl 1988, Yedidia et al. 2000)
        parallel and sequential min-sum and max-product message passing (also for higher-order models)
        message damping (Wainwright 2008)
    Tree-reweighted Belief Propagation (TRBP) (Wainwright et al. 2005)
        parallel and sequential min-sum and max-product message passing (also for higher-order models)
        message damping (Wainwright 2008)
    A-star branch-and-bound search (Bergtholdt et al. 2009)
    Dual Decomposition
        With sub-gradient methods (Kappes et al. 2010)
        With bundle methods (Kappes et al. 2012)
        Automated decomposition of arbitrary factor graphs
        Arbitrary sub-solvers via templates
    Graph Cut (Boykov et al. 2001).
        Push-Relabel (Goldberg and Tarjan 1986)
        Edmonds-Karp (Edmonds and Karp 1972)
        Kolmogorov (Boykov and Kolmogorov 2004)
    QPBO
    MQPBO
    Linear Programming Relaxations over the Local Polytope
    TRWS
    ADSAL
    CombiLP
    Integer Linear Programming
    Multicut (Kappes et al. 2011)
    Fusion Moves for Correlation clustering (Beier et al. 2015)
    Cut & Glue and Cut (Beier et al. 2014)
    Reduced-Inference (Kappes et al. 2013)
    Alpha-Expansion
    Alpha-Beta-Swap
    Alpha-Fusion
    Inf and Flip
    Iterated Conditional Modes (ICM) (Besag 1986)
    Lazy Flipper (Andres et al. 2010)
    Kerninghan Lin
    MCMC Metropolis-Hastings algorithms (Metropolis et al. 1953)
        Gibbs sampling (Geman and Geman 1984)
        Swendsen-Wang sampling (Swendsen and Wang 1987)
    Wrappers around other graphical model libraries
        MRF-LIB
        LIB-DAI
        TRW-S
        QPBO
        GCO
        FastPD
        AD3
        DAOOPT
        MPLP, MPLP-C
Binary HDF5 file format
Command Line Optimizer with built-in protocol mode for runtime and convergence analyses
Python Module with OpenGM C   API exported to Python with boost::python
    Allmost the complete C   API is exported to Python
    Allmost all C   inference algorithms wrapped to Python
    Vectorized API to add multiple functions and factors at once
    Add functions via numpy ndarrays
    Add functions via all default opengm function types
    Extendibility through interfaces for
        custom pure python function types
        custom pure python visitor for inference
            visualization of current inference state with matplotlib 
    Visualize Factor Graph (needs networkx and graphviz)
High performance
    Graphical models with more than 10,000,000 factors
    Specialized functions for optimized cache usage
Extendibility through interfaces for
    custom algorithms
    custom functions
    custom label spaces

@Article{opengm-library,
  author    = {Andres, B. and Beier, T. and Kappes, J.H.},
  title     = {{OpenGM}: {A} {C  } Library for Discrete Graphical Models},
  journal   = {CoRR},
  volume    = {abs/1206.0111},
  year      = {2012},
  url       = {http://arxiv.org/abs/1206.0111}
}
@Article{opengm-benchmark,
  Title                    = {A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems},
  Author                   = {Kappes, J\"orgH. and Andres, Bjoern and Hamprecht, FredA. and Schn\"orr, Christoph and Nowozin, Sebastian and Batra, Dhruv and Kim, Sungwoong and Kausler, BernhardX. and Kr\"oger, Thorben and Lellmann, Jan and Komodakis, Nikos and Savchynskyy, Bogdan and Rother, Carsten},
  Journal                  = {International Journal of Computer Vision},
  Year                     = {2015},
  Number                   = {2},
  Pages                    = {155-184},
  Volume                   = {115},
  Doi                      = {10.1007/s11263-015-0809-x},
  ISSN                     = {0920-5691},
  Language                 = {English},
  Publisher                = {Springer US}
}