@PhdThesis{Zimmermann-PhD2008,
  UPDATE  = { 2009-01-27 },
  author =       {Zimmermann, Karel},
  title =        {Fast learnable methods for object tracking},
  year =         {2008},
  pages =        {105},
  school =       {Czech Technical University},
  address =      {Karlovo n{\' a}m. 13, Praha 2},
  day =          {7},
  month =        {November},
  supervisor =   {Mata, Ji{\v r}{\' i} and Svoboda, Tom{\' a}{\v s}},
  keywords =     {Computer vision, Tracking},
  note =         {{PhD Thesis CTU--CMP--2008--09}},
  project =      {1ET101210407, FP6-IST-027787},
  supervisor =   {Matas, Ji{\v r}{\'\i} and Svoboda, Tom{\'a}{\v s} },
  www =          {ftp://cmp.felk.cvut.cz/pub/cmp/articles/zimmerk/zimmerk-phd.pdf},
  annote = {We propose a learning approach to tracking. The learning
    procedure explicitly minimizes the computational complexity of the
    tracking process subject to a user-defined probability of failure
    (loss-of-lock) and precision. In particular, we studied tracking
    methods estimating the object position by a learned linear mapping
    between intensities and motion.  This mapping is called Learned
    Linear Predictor (LLiP). We extended the predictor to the Sequence
    of LLiPs (SLLiP). The proposed learning approach finds the SLLiP
    with the lowest computational complexity of the tracking subject
    to a user predefined range, i.e., region of motions within which
    the sequential predictor operates, and accuracy. Note, that since
    the individual predictor is a special case of the sequential one,
    the optimal sequence is superior to a single predictor.},
}