Posted on 12/18/2018 3:10:11 AM PST by Freeport
Believe it or not, all these faces are fake. They have been synthesized by Nvidia’s new AI algorithm, a generative adversarial network capable of automagically creating humans, cats, and even cars.
The technology works so well that we can expect synthetic image search engines soon — just like Google’s, but generating new fake images on the fly that look real. Yes, you know where that is going — and sure, it can be a lot of fun, but also scary. Check out the video. It truly defies belief:
According to Nvidia, its GAN is built around a concept called “style transfer.” Rather than trying to copy and paste elements of different faces into a frankenperson, the system analyzes three basic styles — coarse, middle, and fine styles — and merges them transparently into something completely new.
Coarse styles include parameters such as pose, the face’s shape, or the hair style. Middle styles include facial features, like the shape of the nose, cheeks, or mouth. Finally, fine styles affect the color of the face’s features like skin and hair.
According to the scientists, the generator is “capable of separating inconsequential variation from high-level attributes” too, in order to eliminate noise that is irrelevant for the new synthetic face.
For example, it can distinguish a hairdo from the actual hair, eliminating the former while applying the latter to the final photo. It can also specify the strength of how styles are applied to obtain more or less subtle effects.
(Excerpt) Read more at tomsguide.com ...
Wow.
You missed the Thrals.
I thought you wanted an impossible image...
It’s like watching the Kardashians life story.
Maybe they can finally “prove” that Michael Cohen was in Prague!
Starfleet’s rules on using the Holodeck for recreating scenes from Caligula are very clear.
“All actors out of work” :
I hear that there is an opening at a certain bar in NYC.
And there will be no need for endless retakes.
Sex with a robot is coming soon.
It’ll be like......marriage....after she’s bored with you
The effects are much different in 3d I imagine.
So much of movies is created inside computers already; pretty much all of a movie will be, soon.
Our progress in 20 years is incredible...
There is no such thing as “a beautiful Michelle Obama,” in this world or the virtual version.
Wouldn't that be a pendulum swing of galactic proportions?
Photographic evidence is inadmissible if it is digital and not produced from a film camera...accompanied by a negative.
Watch the video. They are taking two real photos and blending them together to make a composite photo. It’s impressive because the two pics aren’t in the same pose or orientation but the AI can extract features and apply them to the base pic for the warping effect but it’s a bit limited now. And in the video one of the example people has glasses, you can tell it’s fake every time because the glasses get all warped. So you need two real people to make one fake one, and it’s only for still photos. It’s not quote the same as generating a photo realistic person from scratch ... yet.
It's already all the rage in Japan.
Minuet!
They should be driving flying cars by now!
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.