This library is an easy-to-use implementation of min-sum belief propagation (i.e., max-product belief propagation). The code is meant to be run from within MATLAB and accordingly provides MATLAB wrappers to the C++ sources. An easy interface allows for specifying discrete factor graph models in terms of unary, pairwise and high-order potentials in just a few lines (see demos). Additionally, the code implements the efficient and exact sparse high-order potential message passing algorithm described in Joint 3D Object and Layout Inference from a single RGB-D Image (GCPR 2015). In contrast to other methods, this algorithm allows to specify special states with both negative as well as positive energies with respect to the default state. Thus attractive as well as repulsive high-order potentials can be realized. For loopy graphs, an approximate solution is returned. The quality of the solution depends on the message passing schedule which can be specified (randomized vs. non-randomized, high-order messages first, etc.). Details on the sparse high-order message computation can be found in our GCPR'15 supplementary.

The source code for this project has been tested on Ubuntu 14.04 and Matlab 2013b and is published under the GNU General Public License. If you find this project useful, we would be happy if you cite us:
@inproceedings{Geiger2015GCPR,
  author = {Andreas Geiger and Chaohui Wang},
  title = {Joint 3D Object and Layout Inference from a single RGB-D Image},
  booktitle = {German Conference on Pattern Recognition (GCPR)},
  year = {2015}
}