Advertise here with Carbon Ads

This site is made possible by member support. ❀️

Big thanks to Arcustech for hosting the site and offering amazing tech support.

When you buy through links on kottke.org, I may earn an affiliate commission. Thanks for supporting the site!

kottke.org. home of fine hypertext products since 1998.

πŸ”  πŸ’€  πŸ“Έ  😭  πŸ•³οΈ  🀠  🎬  πŸ₯”

Scientists can make 3D models of celeb faces and make them talk like other people

Ok, this is a little freaky. If you take a bunch of photos of a person and create a 3D simulation of their face (which is already weird but totally possible) and then use video of someone else speaking to control the 3D face simulation, you can recognize the speaker’s facial expressions and gestures in the 3D face. That sounds a little complicated but just watch the short video clip above. You can clearly see George W. Bush’s facial expressions on the faces of Obama, Tom Hanks, and Hillary Clinton…especially when he says “the legislative process can be ugly”. And the reverse is true as well: even with Bush’s facial expressions and voice, the 3D model of Obama looks like Obama. This is a whole new kind of uncanny valley.

See also real-time facial expression reenactment.