Documentation updates

doc/manual.tex

@@ -16,7 +16,6 @@ Manual\\[1cm]}
 
 \author{Marius Muja, mariusm@cs.ubc.ca\\David Lowe, lowe@cs.ubc.ca}
 
-\date{November 26, 2009}
 
 \begin{document}
 
@@ -294,7 +293,8 @@ Constructs a nearest neighbor search index for a given dataset.
 Index(const Matrix<ElementType>& features, const IndexParams& params);
 \end{Verbatim}
 \begin{description}
-\item[features] Matrix containing the features(points) to index. The size of the matrix is $num\_features \times dimensionality$.
+\item[features] Matrix containing the features(points) that should be indexed, stored in a row-major order (one point 
+of each row of the matrix). The size of the matrix is $num\_features \times dimensionality$.
 \item[params] Structure containing the index parameters. The type of index that will be constructed depends on the type 
 of this parameter. The possible parameter types are:
 
@@ -564,7 +564,7 @@ expected by this function are as follows:
 \begin{description}
  \item[dataset, rows and cols] - are used to specify the input dataset of
 points: dataset is a pointer to a $\rm{rows} \times \rm{cols}$ matrix
-stored in row-major order.
+stored in row-major order (one feature on each row)
 \item [speedup] - is used to return the approximate speedup over linear
 search achieved when using the automatic index and parameter configuration
 (see section \ref{sec:flann_build_index})
@@ -824,7 +824,8 @@ following meaning:
 \begin{description}
 
 \item [\texttt{dataset}] is a $d \times n$ matrix containing $n$
-$d$-dimensional points
+$d$-dimensional points, stored in a column-major order (one feature on
+each column)
 
 \item [\texttt{build\_params}] - is a MATLAB structure containing the
 parameters passed to the function.
@@ -1126,7 +1127,8 @@ class containg the following methods:
     This method builds and internally stores an index to be used for future
 nearest neighbor matchings. It erases any previously stored index, so in order
 to work with multiple indexes, multiple instances of the FLANN class must be
-used. The \texttt{dataset} argument must be a 2D numpy array or a matrix. The
+used. The \texttt{dataset} argument must be a 2D numpy array or a matrix, stored in a
+row-major order (a feature on each row of the matrix). The
 rest of the arguments that can be passed to the method are the same as
  those used in the \texttt{build\_params} structure from
 section \ref{sec:flann_build_index}. Similar to the MATLAB version, the index