[Gzz-commits] manuscripts/Paper paper.tex

[Tuomas J. Lukka]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Changes by:     Tuomas J. Lukka <address@hidden>        03/03/31 05:32:22

Modified files:
        Paper          : paper.tex 

Log message:
        One more move of the texture stuff.

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/Paper/paper.tex.diff?tr1=1.79&tr2=1.80&r1=text&r2=text

Patches:
Index: manuscripts/Paper/paper.tex
diff -u manuscripts/Paper/paper.tex:1.79 manuscripts/Paper/paper.tex:1.80
--- manuscripts/Paper/paper.tex:1.79    Mon Mar 31 04:56:51 2003
+++ manuscripts/Paper/paper.tex Mon Mar 31 05:32:22 2003
@@ -264,6 +264,55 @@
 %XXX: resolution-dependency?
 
 
+Psychophysical studies on texture perception have mostly concentrated
+on pre-attentive 
+\emph{visual texture discrimination}\cite{julesz62visualpattern}, 
+the ability of human observers to effortlessly discriminate
+pairs of certain textures (see Bergen\cite{bergen91theories} for a review). 
+Discrimination models can provide insight on the pre-attentive 
+processes underlying visual perception.
+
+Julesz\cite{julesz81textons} proposed that texture discrimination could be 
+explained by the densities of textons, fundamental texture elements, such as
+elongated blobs, line terminators, and line crossings. 
+However, the textons are hard to define formally.
+
+Much simpler filtering-based models can explain texture discrimination
+just as well \cite{bergen88earlyvision}.
+In these models, a bank of linear filters is applied to the texture followed
+by a nonlinearity and then another set of filters to extract features
+(see, e.g., Heeger\cite{heeger95pyramid}).
+There is also physiological evidence of the filtering processes:
+%The first stages 
+%of visual perception 
+%are fairly well known
+in the visual cortex, there are cells sensitive to different 
+frequencies, orientations, and locations in the visual field
+(see, e.g.,~Bruce et al\cite{bruce96visualperception}).
+
+On a higher level, the correlations between local features are combined 
+by forming contours and possibly
+other higher-level constructions (see, e.g., \cite{saarinen97integration}). 
+These higher levels are not yet thoroughly understood;
+some theories
+(see, e.g., Biederman\cite{biederman87})
+assume certain primitive shapes whose 
+structure facilitates recognition.
+
+The simple model we use here assumes
+that at some point,
+the results from the different pre-attentive feature detectors,
+such as different shapes and colors, 
+are combined to form an abstract \emph{feature vector}
+(see Fig.~\ref{fig-perceptual}).
+The feature vector is then used to compute 
+which concept the particular
+input corresponds to by comparing it to memorized models
+in a simple perceptron-like 
+fashion\cite{rosenblatt62neurodynamics,widrow60adaptive}.
+This configuration is commonly used in neural computation.
+
+
 % In this article, we apply texture shading to synthesize a large number
 % of unique textures for distinguishing virtual objects.
 
@@ -519,42 +568,6 @@
 In order to design a distinguishable distribution of textures,
 we have to take into account the properties of the human
 visual system.
-
-Psychophysical studies on texture perception have mostly concentrated
-on pre-attentive 
-\emph{visual texture discrimination}\cite{julesz62visualpattern}, 
-the ability of human observers to effortlessly discriminate
-pairs of certain textures (see Bergen\cite{bergen91theories} for a review). 
-Discrimination models can provide insight on the pre-attentive 
-processes underlying visual perception.
-
-Julesz\cite{julesz81textons} proposed that texture discrimination could be 
-explained by the densities of textons, fundamental texture elements, such as
-elongated blobs, line terminators, and line crossings. 
-However, the textons are hard to define formally.
-
-Much simpler filtering-based models can explain texture discrimination
-just as well \cite{bergen88earlyvision}.
-In these models, a bank of linear filters is applied to the texture followed
-by a nonlinearity and then another set of filters to extract features
-(see, e.g., Heeger\cite{heeger95pyramid}).
-There is also physiological evidence of the filtering processes:
-%The first stages 
-%of visual perception 
-%are fairly well known
-in the visual cortex, there are cells sensitive to different 
-frequencies, orientations, and locations in the visual field
-(see, e.g.,~Bruce et al\cite{bruce96visualperception}).
-
-On a higher level, the correlations between local features are combined 
-by forming contours and possibly
-other higher-level constructions (see, e.g., \cite{saarinen97integration}). 
-These higher levels are not yet thoroughly understood;
-some theories
-(see, e.g., Biederman\cite{biederman87})
-assume certain primitive shapes whose 
-structure facilitates recognition.
-
 % The seed for randomly choosing
 % an easily distinguishable unique background from a
 % distribution based on 
@@ -590,19 +603,6 @@
 
 % The basic assumption of the model is that an image
 % is perceived as a set of features 
-
-The simple model we use here assumes
-that at some point,
-the results from the different pre-attentive feature detectors,
-such as different shapes and colors, 
-are combined to form an abstract \emph{feature vector}
-(see Fig.~\ref{fig-perceptual}).
-The feature vector is then used to compute 
-which concept the particular
-input corresponds to by comparing it to memorized models
-in a simple perceptron-like 
-fashion\cite{rosenblatt62neurodynamics,widrow60adaptive}.
-This configuration is commonly used in neural computation.
 
 This 
 rough, qualitative