IS = { zkontrolovano 24 Jan 2014 },
  UPDATE  = { 2013-09-25 },
  author =      {Ma{\v c}{\'a}k, Jan and Drbohlav, Ond{\v r}ej},
  language =    {English},
  title =       {Towards Learning Hierarchical Compositional Models in the Presence of Clutter},
  year =        {2013},
  pages =       {532-541},
  booktitle =   {Image Analysis and Processing - ICIAP 2013},
  editor =      {Petrosino, Alfredo},
  publisher =   {Springer},
  address =     {Heidelberg, Germany},
  isbn =        {978-3-642-41180-9},
  volume =      {8156},
  series =      {Lecture Notes in Computer Science},
  book_pages =  {858},
  month =       {September},
  day =         {9-13},
  venue =       {CVPRLab of the University of Naples Parthenope, Naples, Italy},
  organization ={The Group of Italian Researchers in Pattern Recognition (GIRPR), IAPR},
  annote =      {Our goal is to identify hierarchical compositional
    models from highly cluttered data. The data to learn from are
    assumed to be imperfect in two respects. Firstly, large portion of
    the data is coming from background clutter. Secondly, data
    generated by a recursive compositional model are subject to random
    replacements of correct descendants by randomly chosen ones at
    every level of the hierarchy.  In this paper, we study the limits
    and capabilities of an approach which is based on likelihood
    maximization. The algorithm makes explicit probabilistic
    assignments of individual data to compositional model and
    background clutter. It uses these assignments to effectively focus
    on the data coming from the compositional model and iteratively
    estimate their compositional structure.},
  keywords =    {graphical models, unsupervised learning},
  prestige =    {international},
  project =     {GACR P103/12/1578},
  doi =         {10.1007/978-3-642-41181-6_54},