/[gzz]/manuscripts/Paper/paper.tex
ViewVC logotype

Diff of /manuscripts/Paper/paper.tex

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

revision 1.13 by jvk, Thu Mar 6 19:06:32 2003 UTC revision 1.14 by jvk, Sat Mar 8 10:12:22 2003 UTC
# Line 247  surface shape\cite{interrante97illustrat Line 247  surface shape\cite{interrante97illustrat
247    
248  \subsection{Texture perception}  \subsection{Texture perception}
249    
250  Studies on texture perception has mostly concentrated on  Studies on texture perception have mostly concentrated on
251  texture discriminability.  texture discrimination,
252    
253  First experiments on computer-generated, unnatural textures  First experiments on computer-generated, unnatural textures
254  in the 60s (see, e.g, \cite{julesz62visualpattern}) lead to  in the 60s (see, e.g, \cite{julesz62visualpattern}) led to
255  proposals of texture discrimation models based on  proposals of texture discrimation models based on
256  $N$th order statistics of pixels and elaborate connectivity  $N$th order statistics of pixels and elaborate connectivity
257  structures of certain micropatterns.XXX  structures of certain micropatterns.XXX
258    
259  Of cource, such models are not directly applicable on high-resolution  Textons \cite{julesz81textons}: elongated blobs, line terminators,
260    line crossings, etc.
261    
262    Of course, such models are not directly applicable on high-resolution
263  textures; some kind of filtering would be required to obtain the input.    textures; some kind of filtering would be required to obtain the input.  
264  However, the filtering itself can also have good explanatory power  However, the filtering itself can also have good explanatory power
265  \cite{bergen88earlyvision}.XXX  \cite{bergen88earlyvision}.XXX
266    
267    Filtering based approach, e.g., \cite{heeger95pyramid}.
268    
269    
270    Statistical modeling of textures as samples from a probability
271    distribution on a random field as already seen in \cite{julesz62visualpattern}
272    in a very simple form.
273    The most popualar approach is Markov random fields
274    \cite{cross83markov, geman84stochastic}, where a texture
275    is characterized by its local statistics.
276    
277  \subsection{Focus+Context views}  \subsection{Focus+Context views}
278    
279  Focus+Context, or, fisheye views\cite{fc-fisheye} are  Focus+Context, or, fisheye views\cite{fc-fisheye} are

Legend:
Removed from v.1.13  
changed lines
  Added in v.1.14

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