/[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 1.95 by tjl, Mon Mar 31 14:38:10 2003 UTC revision 1.96 by tjl, Mon Mar 31 14:52:25 2003 UTC
# Line 294  assume certain primitive shapes whose Line 294  assume certain primitive shapes whose
294  structure facilitates recognition.  structure facilitates recognition.
295    
296    
297    There have also been studies on
298    mapping texture appearance to an Euclidian texture space
299    (see \cite{gurnsey01texturespace} and the references therein):
300    in the reported experiments, three dimensions have been sufficient
301    to explain most of the variation in the similarity judgements for
302    artificial textures.
303    However, the texture stimuli have been somewhat simple
304    (no color, lack of frequency-band interaction, etc.).
305    For some natural texture sets,
306    three dimensions have also been
307    sufficient \cite{rao96texturenaming}, but often semantic connections cause the
308    similarity to be context-dependant, making it hard to assess the
309    dimensionality.
310    %% XXX: this is something we should experiment with our textures
311    
312  % In this article, we apply texture shading to synthesize a large number  % In this article, we apply texture shading to synthesize a large number
313  % of unique textures for distinguishing virtual objects.  % of unique textures for distinguishing virtual objects.
314    
# Line 1483  exactly what the unique background textu Line 1498  exactly what the unique background textu
1498  %which documents the fragments are from.  %which documents the fragments are from.
1499    
1500  \section{Conclusions}  \section{Conclusions}
   
 We have presented a perceptually designed hardware-accelerated  
 algorithm  
   
1501  TJL  TJL
1502    
1503  We have introduced procedurally generated unique backgrounds  We have presented a perceptually designed hardware-accelerated
1504  as a way of visualizing the identity of data or documents.  algorithm for generating recognizably unique backgrounds.
   
1505  The motivating example, the BuoyING user interface demonstrates  The motivating example, the BuoyING user interface demonstrates
1506  that the method is at its most useful  that the method is at its most useful
1507  when the same document can be reached through  when the same document can be reached through
# Line 1501  Of course, we cannot hope to match in qu Line 1511  Of course, we cannot hope to match in qu
1511  appearance designed by a human designer;  appearance designed by a human designer;
1512  magazines and web sites have long used skilfully designed  magazines and web sites have long used skilfully designed
1513  graphical elements to make themselves recognizable.  graphical elements to make themselves recognizable.
1514  However, the algorithm is able to generate an unlimited  However, our algorithm is able to generate an unlimited
1515  amount of unique backgrounds cheaply, making it possible  amount of unique backgrounds cheaply, making it possible
1516  e.g. for all academic articles with similar typography  e.g. for all academic articles with similar typography
1517  to have their own background.  to have their own background.
# Line 1510  to have their own background. Line 1520  to have their own background.
1520    
1521  So far, we have concentrated mostly on low-end hardware, and  So far, we have concentrated mostly on low-end hardware, and
1522  have not even tapped the full potential of the NV25 architecture.  have not even tapped the full potential of the NV25 architecture.
1523  Given that there are 8 register combiner stages and 4 texture units,  
1524  as opposed to 2 and 2 on NV10,  %Given that there are 8 register combiner stages and 4 texture units,
1525  using 2 passes as we currently do is too much;  %as opposed to 2 and 2 on NV10,
1526  it should be possible to obtain interesting textures with just one pass.  %using 2 passes as we currently do is too much;
1527  The more general ARB fragment program extension also ...  %it should be possible to obtain interesting textures with just one pass.
1528    %The more general ARB fragment program extension also ...
1529    
1530  %We are also working on implementing  %We are also working on implementing
1531  %these algorithms on OpenGL ARB extensions ..., due to their recent release  %these algorithms on OpenGL ARB extensions ..., due to their recent release
1532  %of a Linux driver.  %of a Linux driver.
# Line 1538  The more general ARB fragment program ex Line 1550  The more general ARB fragment program ex
1550  % to use repeating units for non-repeating backgrounds --- but use more than one,  % to use repeating units for non-repeating backgrounds --- but use more than one,
1551  % which are not rationally related.  % which are not rationally related.
1552    
1553  We report an initial recognition experiment  We report an initial recognition experiment showing that the textures
1554    generated by our algorithm are indeed recognizable.
1555  Carrying out usability more tests is necessary, both  Carrying out usability more tests is necessary, both
1556  to measure how well textures can be remembered  to measure how well textures can be remembered
1557  and to make the ad hoc distributions more experimentally based.  and to make the ad hoc distributions more experimentally based.
1558    
1559  %It could also be worthwhile to experiment with other ways  %It could also be worthwhile to experiment with other ways
1560  %of visualizing identity, such as unique edge shapes.  %of visualizing identity, such as unique edge shapes.
1561    
1562  XXX: the effect of zoomability on recognizability  % XXX: the effect of zoomability on recognizability
   
 There have been studies on  
 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,  
 three dimensions have also been  
 sufficient \cite{rao96texturenaming}, 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  
1563    
1564  % The new graphics chips, ATI R300 and NVIDIA NV30 support  % The new graphics chips, ATI R300 and NVIDIA NV30 support
1565  % a great deal more of procedural texturing and it will  % a great deal more of procedural texturing and it will

Legend:
Removed from v.1.95  
changed lines
  Added in v.1.96

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