/[gzz]/gzz/Documentation/Manuscripts/Paper/paper.tex
ViewVC logotype

Diff of /gzz/Documentation/Manuscripts/Paper/paper.tex

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.154 by tjl, Sun Dec 1 11:12:18 2002 UTC revision 1.155 by tjl, Sun Dec 1 11:38:38 2002 UTC
# Line 339  by forming contours and possibly Line 339  by forming contours and possibly
339  other higher-level constructions.  other higher-level constructions.
340  These higher levels are not yet thoroughly understood;  These higher levels are not yet thoroughly understood;
341  theories of structural object perception  theories of structural object perception
342  (see e.g. Biederman\cite{biederman87})  (see, e.g., Biederman\cite{biederman87})
343  propose certain primitive shapes whose  assume certain primitive shapes whose
344  structure facilitates recognition.  structure facilitates recognition.
345    
 We use a rough, qualitative model of visual perception  
 (see Fig.~\ref{fig-perceptual}).  
   
346  \begin{figure}  \begin{figure}
347  \centering  \centering
348  %\fbox{\vbox{\vskip 3in}}  %\fbox{\vbox{\vskip 3in}}
# Line 365  that at some point, Line 362  that at some point,
362  the intensities of different features,  the intensities of different features,
363  such as local and global shapes and colors,  such as local and global shapes and colors,
364  combine to form an abstract \emph{feature vector},  combine to form an abstract \emph{feature vector},
365  as used in neural computation.  as used in neural computation
366    (see Fig.~\ref{fig-perceptual}).
367  This feature vector is then used as an input to the  This feature vector is then used as an input to the
368  networks which linearly compute which concept the particular  networks which linearly compute which concept the particular
369  input corresponds to, in a perceptron-like  input corresponds to, in a perceptron-like
370  fashion\cite{widrow60adaptive,rosenblatt62neurodynamics}.  fashion\cite{widrow60adaptive,rosenblatt62neurodynamics}.
371    
372  From this picture, we can formulate the following points:  
373    From this
374    rough, qualitative model of visual perception
375    we can formulate the following points:
376  To be distinguishable, \label{secfeaturevec}  To be distinguishable, \label{secfeaturevec}
377  \begin{itemize}  \begin{itemize}
378  \item A feature vector for a given texture should be always the  \item A feature vector for a given texture should be always the

Legend:
Removed from v.1.154  
changed lines
  Added in v.1.155

savannah-hackers-public@gnu.org
ViewVC Help
Powered by ViewVC 1.1.26