43 |
|
|
44 |
% WHY HW ACCEL NEEDED - PIXELLATION - NO SAVING! |
% WHY HW ACCEL NEEDED - PIXELLATION - NO SAVING! |
45 |
|
|
|
NO SEMANTIC CORRELATIONS!!!!!!! |
|
|
|
|
46 |
TJL |
TJL |
47 |
|
|
48 |
We present a perceptually designed hardware-accelerated |
We present a perceptually designed hardware-accelerated |
582 |
objects, ... XXX SO!!! JVK |
objects, ... XXX SO!!! JVK |
583 |
|
|
584 |
\begin{figure} |
\begin{figure} |
585 |
|
\centering |
586 |
%\fbox{\vbox{\vskip 3in}} |
%\fbox{\vbox{\vskip 3in}} |
587 |
\ifpics |
\ifpics |
588 |
\def\visscal{.6} |
\def\visscal{.65} |
589 |
a)\\\includegraphics[scale=\visscal]{visper.1}\\\\\\ |
%a)\includegraphics[scale=\visscal]{visper.1}\\ |
590 |
b)\\\includegraphics[scale=\visscal]{visper.2} |
%b)% |
591 |
|
\includegraphics[scale=\visscal]{visper.2} |
592 |
\fi |
\fi |
593 |
\caption{ |
\caption{ |
594 |
\label{fig-perceptual} |
\label{fig-perceptual} |
595 |
a) The qualitative model of visual perception used to create |
The qualitative model of visual perception used to create |
596 |
the algorithm. |
the algorithm. |
597 |
The visual input is transformed into a feature vector, |
The visual input is transformed into a feature vector, |
598 |
which contains numbers (activation levels) corresponding to |
which contains numbers (activation levels) corresponding to |
599 |
e.g.~colors, edges, curves and small patterns. |
e.g.~colors, edges, curves and small patterns. |
600 |
The feature vector is matched against the memorized textures |
The feature vector is matched against the memorized textures. |
601 |
with a simple model. |
In order to generate recognizable textures, random seed |
602 |
The simplicity of the pre-attentive feature matcher |
values should produce |
603 |
is the key feature of the model and is used in the text |
a distribution of feature vectors |
604 |
to formulate principles |
with maximum entropy. |
605 |
about recognizability. |
%The simplicity of the pre-attentive feature vector and matcher |
606 |
b) In order to generate recognizable textures, |
%is the key feature of the model and is used in the text |
607 |
Key is also that producing a distribution of feature vectors |
%to formulate principles |
608 |
with maximum entropy is... |
%about recognizability. |
609 |
} |
} |
610 |
\end{figure} |
\end{figure} |
611 |
|
|