Hi @m4dcow from the research I've found, i think its based off of a mix of the photometric scan normals and regular photos. Check out this guys scan breakdown - https://www.artstation.com/artwork/LqWGw Yeah for ramping up to a 3d face scan, it will take a light stage ):
Hey @m4dcow I'm currently working on extracting specular with the cross pol and parallel pol data. Problem is I'm getting blown out areas, is that where the directional lighting becomes necessary (photometric lighting)? I have two lights at 45 degree angles on each side of my camera. Here are my results. I'm having trouble…
@m4dcow So you've got the 8 light setup for normal/height building, those can all be turned on to capture the albedo. In his blog, he ends up adding 4 direct lights shooting straight down that sit around the lens. Those 4 have cross pol gels. I think trying to do both the photometric data and xpol data with one set of…