This is gonna be the thread where we talk about stuff

@Stij: I don't really get it, what is that supposed to accomplish?

- What it does on a basic level: Take pictures of real world (~10-100), get 3D out of it.
- What it does that other things of that description aren’t doing: Treating the images as a collection of light rays to sample the depth of objects.
Why you might want to pursue this instead of those other techniques: this one shows promise to succeed where other methods break down, e.g. laser scanning has trouble reconciling when things are behind other things and can take a lot of cleanup by hand.
- What is the point of doing this: scanned objects have uses in set reconstruction (e.g. blueprints can be misleading, you may need to reconstruct clutter), film post-production (make that vase fly around), and 3D-printing nifty figurines (a little far-fetched, but people already do it), among other things.

They may be weaknesses to this technique, but you never really find a good assessment of this in a paper itself; you usually have to ask an expert in the field why people don’t use something, and even then it may just be a pet theory.

(as a minor acolyte of computer graphics, I do get SIGGRAPH envy, yes)