Real-Time Expression Transfer for Facial Reenactment
graphics.stanford.edu
graphics.stanford.edu
My only gripe is that there is always so much focus on the face. I understand that its subtleties make for a great research topic; however, I'd love to see this research applied to the full body. Given a full body shot, could we apply the same techniques to gait?
One of the major hurdles I see is background estimation if you move a limb (though, this might be able to be resolved with the same techniques used in photoshop for content-aware fill.
David Foster Wallace [I misremembered it as before edit Philip K. Dick] predicated this would happen. First it would be used to make you look like you were paying attention then it would be used to make you more presentable. He then predicts that when everyone does this it will have the effect of everyone abandoning video telepresence altogether and just going back to using voice only. I can't find the quote.
The proposed solution to what the telecommunications industry's psychological consultants termed Video-Physiognmoic Dsyphoria (or VPD) was, of course, the advent of High-Definition Masking. Mask-wise, the initial option of High-Definition Photographic Imaging — i.e. taking the most flattering elements of a variety of flattering multi-angle photos of a given phone-consumer and ‚ thanks to existing image-configuration equipment already pioneered by the cosmetics and law-enforcement industries — combining them into a wildly attractive high-def broadcastable composite of a face wearing an earnest, slightly overintense expression of complete attention.
[1] https://en.wikipedia.org/wiki/Development_of_L.A._Noire#Tech...
I would be more interested to see where we get with synthesis actors. I could see them creating a single recurring screen actor from 3 or more human beings. One or more for appearance. One for particularly good facial expressions. Another for voice. Maybe another for movement.