Introduction

OpenGM is a C++ template library for discrete factor graph models and distributive operations on these models. OpenGM Python exports the functionality of the C++ library OpenGM to Python. It includes state-of-the-art optimization and inference algorithms beyond message passing. OpenGM handles large models efficiently, since functions that occur repeatedly need to be stored only once . No restrictions are imposed on the factor graph or the operations of the model. The graphical model data structure, inference algorithms and different encodings of functions interoperate through well-defined interfaces. The binary OpenGM file format is based on the HDF5 standard and incorporates user extensions automatically.

The mathematical foundation can be found in the OpenGM Manual .

Contact

opengm (AT) hci.iwr.uni-heidelberg (DOT) de

Authors

OpenGM Authors

  • Bjoern Andres
    • bjoern (AT) andres (DOT) sc
    • www.andres.sc
    • Bjoern Andres,SEAS, Harvard University
  • Thorsten Beier
    • thorsten.beier (AT) iwr (DOT) uni-heidelberg (DOT) de
    • Thorsten Beier, Multidimensional Image Processing Group Heidelberg Collaboratory for Image Processing (HCI), University of Heidelberg
  • Joerg H. Kappes
    • kappes (AT) math (DOT) uni-heidelberg (DOT) de
    • ipa.iwr.uni-heidelberg.de/jkappes
    • Joerg Hendrik Kappes, Image & Pattern Analysis Group Heidelberg Collaboratory for Image Processing (HCI), University of Heidelberg

OpenGM Python Author(s)

  • Thorsten Beier
    • thorsten.beier (AT) iwr (DOT) uni-heidelberg (DOT) de
    • Thorsten Beier, Multidimensional Image Processing Group Heidelberg Collaboratory for Image Processing (HCI), University of Heidelberg