Show Posts
1
General / Re: How to capture photos for a piece of kelp?
« on: May 13, 2017, 06:22:32 AM »
Yea, I think your subject is very problematic for many reasons.
1) The surface is shiny, so the reflection will change at different angles thereby not allowing the software to identify fixed references
2) The object is soft, so there is likely to be movement between shots
3) The shapes are highly variegated, causing occlusion of surfaces at almost every angle
I think this software can do a lot of amazing things, but the type of form you are working with is near impossible without some serious equipment or an extremely creative process.
The main concept you need to keep in mind is that in order to gather 3D data from 2D sources you need to triangulate points, meaning that a point needs to be easily identifiable from at the very least 2 angles.
If you identify a point in one picture, but in the next picture it is blocked by something, the surface color changes, or it moves relative to the surface, if any of these things happen, then the capture will be flawed.
The only way I could think of capturing an organic form as complicated as seaweed you would first need to harden it so it doesn't move, or devise a rig that allows you to shoot at all angles without touching it. Second you would have to coat the surface with some kind of matte coating. And lastly you would need to take an extreme amount of photos in order to account for all the crevices/occlusion.
1) The surface is shiny, so the reflection will change at different angles thereby not allowing the software to identify fixed references
2) The object is soft, so there is likely to be movement between shots
3) The shapes are highly variegated, causing occlusion of surfaces at almost every angle
I think this software can do a lot of amazing things, but the type of form you are working with is near impossible without some serious equipment or an extremely creative process.
The main concept you need to keep in mind is that in order to gather 3D data from 2D sources you need to triangulate points, meaning that a point needs to be easily identifiable from at the very least 2 angles.
If you identify a point in one picture, but in the next picture it is blocked by something, the surface color changes, or it moves relative to the surface, if any of these things happen, then the capture will be flawed.
The only way I could think of capturing an organic form as complicated as seaweed you would first need to harden it so it doesn't move, or devise a rig that allows you to shoot at all angles without touching it. Second you would have to coat the surface with some kind of matte coating. And lastly you would need to take an extreme amount of photos in order to account for all the crevices/occlusion.
