this is for holding javascript data
Volker Strobel edited DeclareMathOperator_argmin_arg_min_DeclareMathOperator__1.tex
almost 8 years ago
Commit id: 906601f66dccdd149ea7783a0d34a36f6f0c2bb5
deletions | additions
diff --git a/DeclareMathOperator_argmin_arg_min_DeclareMathOperator__1.tex b/DeclareMathOperator_argmin_arg_min_DeclareMathOperator__1.tex
index 47f2852..0a1df1a 100644
--- a/DeclareMathOperator_argmin_arg_min_DeclareMathOperator__1.tex
+++ b/DeclareMathOperator_argmin_arg_min_DeclareMathOperator__1.tex
...
used. Therefore, the training dataset will not show any variance, if
the same images were used.
\subsection{Texton Dictionary Generation}
\label{sec:text-dict-gener}
For generating a suitable dictionary for a given environment, an
initial flight was performed. During this flight, 1000 randomly
selected image patches of size $w \times h = 6 \times 6$ were
extracted per image. In total, 100 images were used, resulting in
$100,000$ image patches, which were clustered to create a texton
dictionary. The resulting cluster centers---the prototypes of the
clustering result---are the textons~\cite{varma2003texture}. An
example of a learned dictionary can be found in
Figure~\ref{fig:dictionary}.
Different situations require different textons and a different number
of them. The choice of these parameters is map-dependent, and we set
it to 20 textons for all maps.
% TODO: Write some stuff about the low-level implementation