PDA

View Full Version : How to measure reflectivity?


CHRiTTeR
03-09-2009, 06:53 PM
I guess the thread title says it all.

How does one measure correct reflectivity of a material? Including the fallof curve etc.
Glossyness would be handy too :D

I know there is probably some special expensive equipment for this, but im wondering/hoping if there is any way to do this in a relative cheap way?

noouch
03-09-2009, 07:38 PM
A gonioreflectometer would be the best way to do it:
http://www.graphics.cornell.edu/research/measure/gonio.html

Otherwise, some of the methods used here should be useful:
http://gl.ict.usc.edu/Research/FaceScanning/

Specifically, the difference between the images with and without polarizing filters yields the specular contribution to the image.

Debevec's light stage method goes a long way (I've seen it produce beautiful results applied to various things other than faces). The downside is that you'll probably have to write your own shader, and the data sets can be quite large.

If you just want to measure the glossiness at a single point, things get a lot easier: Build a pinhole light source (cover a lamp with aluminum foil and poke a tiny hole for light to get through. Set up a camera on a tripod (it should have manual exposure control so you can expose long enough to see the result). Then turn on the lamp, align it so that the pinhole of light reflects from the desired point, and take a picture. From the image, you should be able to deduce the approximate width of the highlight.

In short: measuring reflectance across an entire object is a lengthy process, and you're probably going to need calibrated equipment and an elaborate rig for it. That's not to say you should be discouraged, no doubt anything you build in this direction will be really cool and score you serious geek cred ;)

mister3d
03-09-2009, 07:44 PM
Specifically, the difference between the images with and without polarizing filters yields the specular contribution to the image.


I might be wrong, but polarizing filters remove just a part of reflection? There is a polarized reflection and not polarized?

noouch
03-09-2009, 07:46 PM
I might be wrong, but polarizing filters remove just a part of reflection? There is a polarized reflection and not polarized?
If you polarize the light source in the right direction and put another polarizing filter in front of the lens, it will completely remove all specular reflection while maintaining the diffuse portion. I think this only works with linear polarizers though...

CGTalk Moderation
03-09-2009, 07:46 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.