Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
AI Education

Using AI and Photoshop to Fake 'Photos' of Ancient Roman Emperors (theverge.com) 25

Machine learning "can even bring ancient statues to life, transforming the chipped stone busts of long-dead Roman emperors into photorealistic faces you could imagine walking past on the street," reports the Verge, citing a new project by a film-industry VR specialist.

Slashdot reader shirappu summarizes their report: Daniel Voshart's work on creating life-life images of Roman emperors from their statues started as a quarantine project and quickly got out of hand. His portraits of the emperors (a collection of 54 as of July) are created using generative adversarial networks, which are fed images of the emperors from statues, coins, and paintings. These are then edited and tweaked based on historical descriptions, and reworked in PhotoShop, where Voshart says he can "avoid falling down the path into uncanny valley."

Voshart has written about the process himself here.

The Verge writes: To help, he says he sometimes fed high-res images of celebrities into the GAN to heighten the realism. There's a touch of Daniel Craig in his Augustus, for example, while to create the portrait of Maximinus Thrax he fed in images of the wrestler André the Giant... The process, as he describes it, is almost alchemical, relying on a careful mix of inputs to create the finished product...

What's more, his work is already enticing academics, who have praised the portraits for giving the emperors new depth and realism. .. As a sort of thank you to his advisers, Voshart has even used a picture of one USC assistant professor who looks quite a bit like the emperor Numerian to create the ancient ruler's portrait.

This discussion has been archived. No new comments can be posted.

Using AI and Photoshop to Fake 'Photos' of Ancient Roman Emperors

Comments Filter:

It is easier to write an incorrect program than understand a correct one.

Working...