About Me

The Mona Lisa came back to life with artificial intelligence

The Mona Lisa came back to life with artificial intelligence

A new Amnesty International experience saw the emergence of the Mona Lisa's life. Famous artwork is just a collection of symbolic faces that have recently appeared in fake videos.

Based on one face image, researchers from the Samsung AI Center in Russia have developed a new algorithm that can create this kind of fake footage.


More than 7,000 YouTube videos have been fed into the system to create the algorithm. This means that he was able to learn the movements of the "landmarks" in the face.

By observing the forms of the mouth, eyes and nose of the face, the algorithm can then apply the same movements to a still image later.

Researchers in Moscow also made similar effects on the images of David Beckham and Marilyn Monroe, where they basically painted their features.


The Mona Lisa brought to life definitely attention ...
Amnesty International can now turn the Mona Lisa into a real person. All with just one picture.

"It carries out lengthy instruction on a wide range of videos," the researchers said in their study published on ARXiv.

"After that, they were able to frame a few - and learn once for the models of the heads of the nervous talk of previously invisible people ...

That's why the algorithm was fed with images of the likes of Mona Lisa and Albert Einstein. Although clearly fake, the final shots look surprisingly realistic.


TechCrunch report that "the quality and clarity of the fake talking head vary widely" in the stills. "The usual swabs and exotic artifacts" are also everywhere, he says.

However, her report adds that "nevertheless, it's great that it works as well as it does." The TechCrunch report, like others, indicates that this new AI only works on the face and upper torso. The Mona Lisa can dance or clap with hands (but maybe "after" is the operative word ...)


Final thoughts

What is your opinion? Amnesty International has divided these views since their release. The mirror suggests that the new algorithm can contain "a number of useful applications," especially in filmmaking.

However, he also acknowledges that artificial intelligence (in the wrong hands) can be misused to create "deep, fake" videos. Last year, several videos of this genre appeared on the Internet, apparently showing celebrities in obvious situations.

It also increases the chances that videos will be shown to politicians, or even people who have died, saying and doing things they have never done.

The telegraph, which also conveys the story, pointed to fears as well. US lawmakers have warned that the fake videos "could harm national security," the report said.

Therefore, as far as this kind of artificial intelligence sexy and fun, in the wrong hands, the consequences can not be conceived. What do you think? Let us know!

Post a Comment

0 Comments