The overall algorithm we presented takes two pieces of data for its input. The first
  is a DEM, with an example given in Figure~\ref{fig:dem_small}. The second is a
  GIS layer, presented in Figure~\ref{fig:shapes}. The algorithm combines its
  input and produces a set of three-dimensional triangles. The algorithm was
  implemented as a preprocessing stage for our visualization tool.

  \begin{figure}[!ht]
    \begin{center}
      \includegraphics[width=3.0in]{images/bw2DShapeFileShadow.eps}
    \end{center}
    \caption{Two-dimensional GIS Layer\label{fig:shapes}}
  \end{figure}

  With our data in hand, we constructed an interactive visualization tool. This
  tool provides better interaction with GIS data than current packages because
  of the ability to accentuate specific features of the data, and because unlike
  other packages you can interact with and manipulate the data. Using our tool,
  one can accentuate the data by stretching, in various directions, the data
  set. This, combined with the ability to become immersed in the data in three
  dimensions and the capability to ``fly through'' the data and view the data
  from multiple angles, provides the user a greater ability to discover features
  that would have otherwise been missed. A screenshot of the visualization is
  shown in Figure~\ref{fig:dem_and_gis}.

  \begin{figure}[!ht]
    \begin{center}
      \includegraphics[width=3.0in]{images/dem&ShapesWhiteBGShadow.eps}
    \end{center}
    \caption{Three-dimensional visualization of combined DEM and GIS layer.  Outlining the
             triangles helps accentuate the height.\label{fig:dem_and_gis}}
  \end{figure}