I was looking up ways to render vector with meshes some time ago and people seem to point to this Microsoft Research paper by Charles Loop and Jim Blinn (yes, that Blinn):
http://research.microsoft.com/en-us/um/people/cloop/loopblinn05.pdf
But this paper is from 2005, and I can't see fruits of this work anywhere in games. I wonder if anything ever came out of it.
Shouldn't all modern game engines have this feature by now? Is it a patented algorithm or something? Is this what mobile OS vector graphics libraries use?