IS = { zkontrolovano 20 Jan 2015 },
  UPDATE  = { 2015-01-20 },
  author =       {Derner, Erik and Svoboda, Tom{\'a}{\v s}},
  title =        {Indexing Images for Visual Memory by Using {DNN} Descriptors
                  -- Preliminary Experiments},
  institution =  {Center for Machine Perception, K13133 FEE Czech Technical
  address =      {Prague, Czech Republic},
  year =         {2014},
  month =        {December},
  type =         {Research Report},
  number =       {CTU--CMP--2014--25},
  issn =         {1213-2365},
  pages =        {16},
  figures =      {11},
  authorship =   {50-50},
  psurl =        {[Derner-TR-2014-25.pdf]},
  project =      { FP7-ICT-609763, SGS13/142/OHK3/2T/13},
  annote =       {Visual memory in mobile robotics is important to make the
                  local- ization of a robot robust to situations, when GPS or
                  similar localization methods are not available. Unlike many
                  conventional approaches us- ing local features, we use a
                  holistic method that employs deep neural networks (DNNs) to
                  calculate a global descriptor of the whole image. We
                  consider a scenario in which a robot equipped with an omni-
                  directional camera calculates and stores DNN descriptors of
                  images together with the positions as it moves in the
                  environment. When the position is unknown to the robot, the
                  algorithm estimates it for a given omnidirectional image by
                  matching it with the most similar database image. We
                  compared our approach with a recently tested GIST-based ap-
                  proach on the same dataset and we found out that the
                  DNN-based approach yields better results. The experiments
                  also show that the DNN-based algorithm is quite robust to
                  partial occlusion, rotation and changes in lighting
  keywords =     {Image indexing, visual localization, deep neural networks},
  comment =      { },