[Gzz-commits] manuscripts ./gzigzag.bib Paper/paper.tex

[Janne V. Kujala]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Changes by:     Janne V. Kujala <address@hidden>        03/03/08 05:12:22

Modified files:
        .              : gzigzag.bib 
        Paper          : paper.tex 

Log message:
        more texture perception

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/gzigzag.bib.diff?tr1=1.59&tr2=1.60&r1=text&r2=text
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/Paper/paper.tex.diff?tr1=1.13&tr2=1.14&r1=text&r2=text

Patches:
Index: manuscripts/Paper/paper.tex
diff -u manuscripts/Paper/paper.tex:1.13 manuscripts/Paper/paper.tex:1.14
--- manuscripts/Paper/paper.tex:1.13    Thu Mar  6 14:06:32 2003
+++ manuscripts/Paper/paper.tex Sat Mar  8 05:12:22 2003
@@ -247,19 +247,32 @@
 
 \subsection{Texture perception}
 
-Studies on texture perception has mostly concentrated on 
-texture discriminability. 
+Studies on texture perception have mostly concentrated on 
+texture discrimination, 
 
 First experiments on computer-generated, unnatural textures 
-in the 60s (see, e.g, \cite{julesz62visualpattern}) lead to 
+in the 60s (see, e.g, \cite{julesz62visualpattern}) led to 
 proposals of texture discrimation models based on 
 $N$th order statistics of pixels and elaborate connectivity 
 structures of certain micropatterns.XXX
 
-Of cource, such models are not directly applicable on high-resolution
+Textons \cite{julesz81textons}: elongated blobs, line terminators,
+line crossings, etc.
+
+Of course, such models are not directly applicable on high-resolution
 textures; some kind of filtering would be required to obtain the input.  
 However, the filtering itself can also have good explanatory power
 \cite{bergen88earlyvision}.XXX
+
+Filtering based approach, e.g., \cite{heeger95pyramid}.
+
+
+Statistical modeling of textures as samples from a probability 
+distribution on a random field as already seen in \cite{julesz62visualpattern}
+in a very simple form.
+The most popualar approach is Markov random fields
+\cite{cross83markov, geman84stochastic}, where a texture 
+is characterized by its local statistics.
 
 \subsection{Focus+Context views}
 
Index: manuscripts/gzigzag.bib
diff -u manuscripts/gzigzag.bib:1.59 manuscripts/gzigzag.bib:1.60
--- manuscripts/gzigzag.bib:1.59        Thu Mar  6 09:59:38 2003
+++ manuscripts/gzigzag.bib     Sat Mar  8 05:12:13 2003
@@ -2598,9 +2598,33 @@
     title = "Visual pattern discrimination",
     journal = "IRE Transactions on Information Theory",
     volume = "8", 
-    issue = "2",
+    number = "2",
     pages = "84--92", 
     year = "1962",
+}
+
address@hidden ------------------
address@hidden MRF texture models
address@hidden ------------------
+
address@hidden cross83markov,
+    author = "Cross, G. C. and Jain, A. K.",
+    title = "Markov random field texture models",
+    journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
+    volume = "5",
+    number = "1",
+    pages = "25--39",
+    year = "1983",
+}
+
address@hidden geman84stochastic,
+author = {Stuart Geman and Donald Geman},
+year = 1984,
+title = {Stochastic relaxation, {G}ibbs distributions, and 
+         the {B}ayesian restoration of images},
+journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
+volume = 6,
+pages = {721--742}
 }
 
 @comment -------------------------------------