BRTDFunc material tutorial?

Hi folks,

I'm trying to do some simple tests within Radiance of material parameters and their appearance and I was wondering if anyone has a good explanation of the BRTDFunc and how one goes about structuring a .cal file for it, etc.

Thanks,

kirk

···

------------------------------

Kirk L. Thibault, Ph.D.
[email protected]

p. 215.271.7720
f. 215.271.7740
c. 267.918.6908

skype. kirkthibault

Interesting site:

http://www1.cs.columbia.edu/CAVE/curet/

is this kind of database directly applicable to a BRTDFunc?

···

------------------------------

Kirk L. Thibault, Ph.D.
[email protected]

p. 215.271.7720
f. 215.271.7740
c. 267.918.6908

skype. kirkthibault

On Jan 27, 2006, at 1:02 PM, Kirk Thibault wrote:

Hi folks,

I'm trying to do some simple tests within Radiance of material parameters and their appearance and I was wondering if anyone has a good explanation of the BRTDFunc and how one goes about structuring a .cal file for it, etc.

Thanks,

kirk

------------------------------

Kirk L. Thibault, Ph.D.
[email protected]

p. 215.271.7720
f. 215.271.7740
c. 267.918.6908

skype. kirkthibault

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

Hi Kirk,

Probably the simplest example using the BRTDfunc type is the glaze.csh script output and its associated glaze1.cal and glaze2.cal auxiliary files. These may be found in the ray/src/gen directory of the latest Radiance distribution. If this is not enough for you, you can look at the attached He.cal file for ideas.

It might be possible to make use of the Curet database, but no one has done it yet with Radiance, as far as I know. You would probably want to apply the plasdata or transdata primitives, rather than BRTDfunc, which is awkward to use with a large dataset directly.

I'm sorry I don't have any better examples of BRDF use, but the common method is to fit the data to one of the Gaussian distribution models, which offer more complete and efficient ray interactions. Specifically, the BRDF types do not account for indirect specular components, though I believe there is some support for this in the photon map extension.

At the 2005 Eurographics Symposium on Rendering, Addy Ngan presented the following paper:

Experimental Analysis of BRDF Models
Addy Ngan, FrŽdo Durand, and Wojciech Matusik

which is online at:

  http://people.csail.mit.edu/addy/research/brdf/

The "supplemental document" linked above contains a very nice list of BRDF materials and their fits to the "Ward model," which is used in Radiance. If you are applying this to a recent version of Radiance (3.6 or later), then you'll actually want to use the coefficients for the "Ward-Duer model," as it includes the correction to the normalization of my original model, pointed out by Arne Duer of the Univ. of Innsbruck (and also Don Walters of Boeing Corporation). Unfortunately, Ngan et al. only cover isotropic materials in this document. They did measure a couple of anisotropic materials, and I have further information on those, but there isn't really enough and the fit wasn't good enough to make them worth using in Radiance.

I hope this helps.
-Greg

He.cal (4.79 KB)