|
agent47crows (November 30, 1999 at 12:00 am)
Well, just google Paul Debevec, he's done quite a lot on 3D, even made an ATI demo on live HDRi.
But yea, col, bump, spec maps can indeed be collected with simple photos.
I am currently working on a 3D scanner. It needs a laser line and a webcam.
I just need the laser line, I have a laser point so far, and I cannot find the line in here, so I have to study refractions.
Imhotep397 (November 30, 1999 at 12:00 am)
Wow...I mean really wow. I never knew normals could be 1.) Calculated from specular and diffuse data 2.) I never knew normal data could be harvested from photographic information and definitely didn't know it could be harvested in real-time. This should be FAR cheaper than Contour's Mova system and has the potential to replace marker based full body tracking and it should be dirt cheap, relatively speaking. Will these guys be showing at Siggraph? Where can I get more info on this?
agent47crows (November 30, 1999 at 12:00 am)
Yea, all that laser stuff is for deep model scanning. By their explanations, I belive they use normal cameras with specific lights and they gather normal (parralax) and diffuse maps which result in a very fast, realtime preview with those enviro. mappings.
QuantumAnomaly (November 30, 1999 at 12:00 am)
Those Env-Mapped renderings are badass!
I was under the impression that people couldn't have their faces scanned with eyes open because of the lasers. I guess they must be using something else. Fascinating stuff...
agentstona (November 30, 1999 at 12:00 am)
now thats whats rendering is about awesome
sjoerdkoala12 (November 30, 1999 at 12:00 am)
cool |
