445 |
|
|
446 |
%XXX: shorten by one half column |
%XXX: shorten by one half column |
447 |
|
|
|
XXX: - visual discrimination experiments: \cite{julesz62visualpattern} |
|
|
|
|
|
XXX: simple models (filtering) can have good explanatory power |
|
|
on texture discrimination\cite{bergen88earlyvision}. |
|
|
|
|
448 |
In this section, we discuss the methods to generate unique |
In this section, we discuss the methods to generate unique |
449 |
background textures on an abstract level. |
background textures on an abstract level. |
450 |
|
|
1120 |
we start from a fairly light background (>80 XXX). |
we start from a fairly light background (>80 XXX). |
1121 |
On a darker background, this approach would probably look silly. |
On a darker background, this approach would probably look silly. |
1122 |
|
|
1123 |
\subsection{Recognizability and memorizability} |
\subsection{Texture perception} |
|
|
|
|
% XXX: shorten by half |
|
|
|
|
|
JVK |
|
1124 |
|
|
1125 |
Psychophysical studies on texture perception have mostly concentrated |
Psychophysical studies on texture perception have mostly concentrated |
1126 |
on \emph{visual texture discrimination}\cite{julesz62visualpattern}, |
on \emph{visual texture discrimination}\cite{julesz62visualpattern}, |
1163 |
|
|
1164 |
In our application visual texture discrimination is not as |
In our application visual texture discrimination is not as |
1165 |
much of an issue as memorizability and recognizability of |
much of an issue as memorizability and recognizability of |
1166 |
previously seen textures. |
previously seen textures, because the textures are ... discriminable.. |
1167 |
|
|
1168 |
|
|
1169 |
|
There have been studies on |
1170 |
|
mapping texture appearance to an Euclidian texture space |
1171 |
|
(see \cite{gurnsey01texturespace} and the references therein): |
1172 |
|
in the reported experiments, three dimensions have been sufficient |
1173 |
|
to explain most of the variation in the similarity judgements for |
1174 |
|
artificial textures. |
1175 |
|
However, the texture stimuli have been somewhat simple |
1176 |
|
(no color, lack of frequency-band interaction, etc.). |
1177 |
|
For some natural texture sets (see, e.g., \cite{rao96texturenaming}), |
1178 |
|
three dimensions have also been |
1179 |
|
sufficient, but often semantic connections cause the |
1180 |
|
similarity to be context-dependant, making it hard to assess the |
1181 |
|
dimensionality. |
1182 |
|
% XXX: this is something we should experiment with our textures |
1183 |
|
|
1184 |
|
\subsection{Recognizability and memorizability} |
1185 |
|
|
1186 |
The textures created by our algorithm, |
The textures created by our algorithm, |
1187 |
although repeating, are more like complete images |
although repeating, are more like complete images |
1188 |
than the statistical microstructure studied in most texture perception work. |
than the statistical microstructure studied in most texture perception work. |
1223 |
have shown that file popularity approximately follows Zipf's law |
have shown that file popularity approximately follows Zipf's law |
1224 |
(see, e.g. [xxxref]) so that a small number of documents accounts |
(see, e.g. [xxxref]) so that a small number of documents accounts |
1225 |
for most of the use. |
for most of the use. |
1226 |
Furthermore, because the texture appearance has no semantic connection |
|
1227 |
with the document, |
Furthermore, because the texture appearance has no correlation |
1228 |
|
with the document content, |
1229 |
the textures of any two important files are similar only by chance. |
the textures of any two important files are similar only by chance. |
1230 |
|
|
1231 |
%Our experience shows that at least the most recurring textures |
%Our experience shows that at least the most recurring textures |
1232 |
%are not only recognized but can |
%are not only recognized but can |
1233 |
%also easily be associated with the document content. |
%also easily be associated with the document content. |
1234 |
|
|
|
XXX |
|
|
|
|
|
Mapping texture appearance to an Euclidian texture space |
|
|
(see \cite{gurnsey01texturespace} and the references therein): |
|
|
in the reported experiments, three dimensions have been sufficient |
|
|
to explain most of the variation in the similarity judgements for |
|
|
artificial textures. |
|
|
However, the texture stimuli have been somewhat simple |
|
|
(no color, lack of frequency-band interaction, etc.). |
|
|
For some natural texture sets (see, e.g., \cite{rao96texturenaming}), |
|
|
three dimensions have also been |
|
|
sufficient, but often semantic connections cause the |
|
|
similarity to be context-dependant, making it hard to assess the |
|
|
dimensionality. |
|
|
% XXX: this is something we should experiment with our textures |
|
|
|
|
1235 |
XXX: refs? |
XXX: refs? |
1236 |
|
|
1237 |
%\section{Software availability} |
%\section{Software availability} |