[passivate]

Specifically, what is 'state of the art' in the iphones image pipeline? Many of these computational techniques (focus stacking, image averaging, subject recognition, face recognition, etc) have been in use for decades - but just not in a handheld device.

By the same token, I wouldn't call Google docs as 'state of the art' since its duplicating existing desktop software, but in a browser (not to minimize their effort, I'm sure its hard).

[packetslave]

You said it yourself: "just not in a handheld device" and not in what is effectively realtime.