Quite a striking result. Given a camera mounted in one’s glasses, pointed at one’s face, this system can synthesize the view of a camera in *front* of you, pointed at you. You could video chat while walking around without holding anything in front of you. youtube.com/watch?v=atzPvW
Conversation
Replying to
The video is a great overview, but lots more available from the authors:
gvv.mpi-inf.mpg.de/projects/EgoCh
17
Replying to
that’s crazy to me. There must be more behind the scenes bc the hair is invisible in the input but perfectly generated in the outputs
1
It requires per-subject training, which presumably conditions the hair.
1
1
13
Show replies
Replying to
I think this is a great innovation which will enable a lot of people to have much much more sociable and happy lives.
I also think that if it gets anywhere over the next few years, it will be talked about as "Now the techbros want to mount a camera on your face"
1
4
Indeed—and I mean, there are of course many hideous implications to tech such as this. And they’re worth thinking about. But we all play the hand we are dealt, and my card says Earnest Twitter.
11
Maybe I should watch the 6-minute “Two Minute Paper” first
1
Show replies
It is going to be used for eye tracking with advertising even if marketed as video chat device.
1
Replying to
It's pretty clever. I guess my only gripe with it is the fact that we'd be losing the non-verbal communication we do with hands.
1
If we can do this trick with faces, where we’re evolutionarily attuned to the tiniest distinctions, it’ll be easy to do it with a couple of cameras on either side of your head pointed at your hands. Could probably even do the whole body that way.
1