May I ask what university Vidmaster?
What are they using to extract objects from the point cloud? Object Oriented classification approaches or something else?
Here in Germany, not in the US.
Saddly, graphics are not mine line of work, so I can't tell you what exactly they are using.
I know there is a project going on with Object Recognition but that's only one of the researcher's favorite project he keeps telling everybody about :-)
I prefer information systems, computer linguistics and various other tasks that have less to do with math and more with smart and clever design philosophies. I've only seen what my colleges are doing and how incredible that looks. I am not the guy too ask.
Still, it's in development. It just seems these guys were faster ;-) And I am pretty sure they are not showing us everything yet, as I have stated before.
I heared they have already registered the patents.