Submitted via IRC for SoyCow8093
The rate at which deepfake videos are advancing is both impressive and deeply unsettling. But researchers have described a new method for detecting a "telltale sign" of these manipulated videos, which map one person's face onto the body of another. It's a flaw even the average person would notice: a lack of blinking.
Researchers from the University at Albany, SUNY's computer science department recently published a paper titled "In Ictu Oculi: Exposing AI Generated Fake Face Videos by Detecting Eye Blinking." The paper details how they combined two neural networks to more effectively expose synthesized face videos, which often overlook "spontaneous and involuntary physiological activities such as breathing, pulse and eye movement."
Source: https://gizmodo.com/most-deepfake-videos-have-one-glaring-flaw-1826869949
(Score: 2) by takyon on Sunday June 17 2018, @05:05AM (2 children)
As long as you match the skin color, pretty much anyone's eyelids could do.
This is a minor blip in Deepfakes' great long journey, which is only beginning. The hardware will get at least an order of magnitude better, and the software will improve as well.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 0) by Anonymous Coward on Sunday June 17 2018, @05:47AM (1 child)
Can they help with plot-lines?
Actually, the ultimate would be to create Deepfake versions of real movies but with sex scenes added. Basically, porn parodies but with the original cast.
(Score: 3, Funny) by Anonymous Coward on Sunday June 17 2018, @12:27PM
The Cosby Show will never be the same again.