Imagine taking one person's body and face, and then being able to make them speak using someone else's voice -- and their mouth movements and facial expressions, too. That's now possible to do in real time, thanks to a research team at Stanford, using only a consumer-grade PC and what appears to be a Kinect depth-of-field camera:
One suggested application for this tech is real-time video of someone speaking in a foreign language, but using the voice and mouth movements of a translator. It can also be used to change the appearance of a voice actor by substituting a new body (think a live-action Shrek), or to add "make-up" or other facial decorations in post-processing. But mostly I think it's going to make me suspicious of the authenticity of live video, in the same way that Photoshop has made me skeptical of every amazing picture I see.
Anyway, that's all for this week. We'll be back on Monday, see you then!
https://www.youtube.com/watch?v=eXVspNUeiWw
Comments