Re: Texel area

Hi all!

I've finally had the time to implement and use all your suggestions, and if
you're interested you can see some of the results here:
http://shadowcovenant.com/blog/2011/02/10/ambient-occlusion-maps/

Most of the issues I solved by using barycentric coordinates, instead of
using just gradients calculated through interpolation...

Thanks for all the help everyone!

Best regards,
Diogo

-----Original Message-----
From: Olivier Galibert [mailto:[email protected]] 
Sent: sexta-feira, 4 de Fevereiro de 2011 19:32
To: [email protected]
Subject: Re: [Algorithms] Texel area

On Fri, Feb 04, 2011 at 10:50:08AM -0000, Diogo de Andrade wrote:
> So, what I want to do is, given
> 
> - triangle T=(V1,V2,V3), in which V1, V2, V3 are vertexes that have 
> some properties (world space position, normal, diffuse color, texture 
> coordinates
> 0 (in uniform space), texture coordinates 1 (in texture space, 
> [0..texture size[), etc),
> 
> - texture size
> 
> - position inside the triangle (absolute value in texture space)
> 
> is to find out the rectangle that bounds the texel in world space 
> (origin+2 vectors)...

If that's your real problem, the solution is reasonably simple and has
nothing to do with the rasterization.

You have three points when space/texture coordinates:
  (x0, y0, z0, u0, v0)
  (x1, y1, z1, u1, v1)
  (x2, y2, z2, u2, v2)

Given u, v the position in texture space, you first want to find the
barycentric coordinates (t1, t2):
 | u0 + t1*(u1-u0) + t2*(u2-u0) = u
 | v0 + t1*(v1-v0) + t2*(v2-v0) = v

That's a rather usual linear system.  Noting the inverse determinant
di:
   di = 1/((v2-v0)*(u1-u0) - (u2-u0)*(v1-v0))

then:
 | t1 = di*((v2-v0)*(u-u0) - (u2-u0)*(v-v0))  | t2 = di*((v1-v0)*(u-u0) -
(u1-u0)*(v-v0))

>From these, you can easily find the world coordinates:
 | x = x0 + t1*(x1-x0) + t2*(x2-x0)
 | y = y0 + t1*(y1-y0) + t2*(y2-y0)
 | z = z0 + t1*(z1-z0) + t2*(z2-z0)

You'll notice that all these are linear systems, which means a texel
rectangle size is independant of its position.  So you can get your vector
once by computing the position differences between (0, 0), (0.5, 0) and (0,
0.5) for instance.  You'll notice the equations simplify nicely when
computing deltas if you do the math.

I suspect it's not your real problem though, so feel free to correct it :-)

  OG.


----------------------------------------------------------------------------
--
The modern datacenter depends on network connectivity to access resources
and provide services. The best practices for maximizing a physical server's
connectivity to a physical network are well understood - see how these rules
translate into the virtual world? 
http://p.sf.net/sfu/oracle-sfdevnlfb
_______________________________________________
GDAlgorithms-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_name=gdalgorithms-list


------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
_______________________________________________
GDAlgorithms-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_name=gdalgorithms-list