\section{Experiment design}

A real-life uncontrolled experiment using the kinect was conducted to
test to the algorithm. 

\subsection{Kinect}

Signals:
\begin{itemize}
\item audio
\item video
\item depth map
\end{itemize}

Skeleton fitting:
\begin{itemize}
\item OpenNI (calibration needed)
\item Microsoft SDK
\end{itemize}

\subsection{Environment}

\begin{itemize}
\item 1 week
\item 23 people
\end{itemize}

\subsection{Data set}

The original dataset consists of the sequence of all the frames where
a skeleton was detected by the Microsoft SDK. For each frames the
following data is available:
\begin{itemize}
\item the 3D coordinates of 20 body joints
\item the z-value: this is the distance from the detected skeleton to
  the camera
\end{itemize}
For some of frames, one or several joints were occluded by another
part of the body. In those cases, the coordinates of these joints are
either absent from the frame or present but tagged as \emph{Inferred}
by the Microsoft SDK. It means that even though the joint was not
present on the frame, the skeleton-fitting algorithm was able to guess
its location.

Each frame also has a skeleton ID number. If this numbers stays the
same across several frames, it means that the skeleton-fitting
algorithm was able to detect the skeleton in a contiguous way. This
allows us to define the concept of a \emph{run}: a sequence of frames
with the same skeleton ID.



%%% Local Variables: 
%%% mode: latex
%%% TeX-master: "kinect"
%%% End: