May 2, 2025
Trending News

How to detect deepfake in video calls

  • August 10, 2022
  • 0

Deepfake. (Photo: Metaphysics) Last month, news broke that a remote interview scam was involved Experience someone else’s personality using deep fakes. And while the purpose of these actions

Deepfake. (Photo: Metaphysics)

Last month, news broke that a remote interview scam was involved Experience someone else’s personality using deep fakes. And while the purpose of these actions has nothing to do with getting a job, it does present a new security issue.

The truth is that, more or less, video calling seems to be here, not only on a professional level, but also on a personal level. Therefore, everyone is subject to deception. The question is: can be easily detected DeepFax? A metaphysical company has developed a curious technique.

A promising and simple tactic to detect deep fakes

This is confirmed by a recent study We “catch” less than half of the deepfakes which pass before our eyes, although, of course, we do not always have the same probability of finding them.

If the shared video is displayed Social networksUnless users have specialized tools, they have no other choice Be attentive and try to determine whether it is a joke or not.

However, if a person is in a video call, The chances of detecting a deep deep fake increase. Another person may be asked to perform gestures that falsify functionality Software which generates a deep fake in real time and thus can be detected when trying to impersonate another person.

team Metaphysic asked actor Bob Doyle to participate in a series of video calls using DeepFaceLive. A fake real-time version of the popular DeepFaceLab software. As you can see from the screenshot, the system handled most of the characters perfectly.

Deepfake. (Photo: Metaphysics)

Doyle happens Sylvester Stallone, Ryan Reynolds, Nicolas Cage, Keanu Reeves, Tom Cruise, Vin Diesel etc. However, in the middle of the show, he was asked to slowly turn his head to the side. Result? The quality of the deep fake is lost when the angle reaches 90 degrees.

As the team explains, most deep fake systems They don’t work well in sharp corner situations.

In fact, almost all criteria are used by the face recognition algorithm They are assigned to the front of the face. So when someone shows his face on a fake profile, he starts to get confused.

Deepfake. (Photo: Metaphysics)

One of the other techniques that metaphysics offers for identifying deep fakes is Ask the person you are video calling with to wave their hand in front of their face. This will create a “serious situation for the model, which can show real-world lag and poor quality overlays.”

as technology As achievements and algorithms are trained more, deep fakes become more accurate and realistic. However, at this time, someone may be asked Turn your head 90 degrees or what Raise your hand to the camera to see if you’re lying.

This AI detects deep fakes on TikTok

A group of researchers from Singapore developed A system artificial intellect Can determine 98.53% of the time whether the analyzed video has been digitally manipulated.

This system was developed by Wang Weimin, a graduate scientist at the National University of Singapore. Currently working ByteDanceparent company TikTokwho participated in a competition with hundreds of applicants organized by the ISGF, a national AI program launched by the National Research Foundation of Singapore.

The competition lasted 5 months and Wang, who after winning received a prize of $300,000 USD to commercialize his idea, assured that In order to fight against manipulation in social networks, he prefers to implement his own technologies in his own company.

Tom Cruise deep fake. (Photo: Awaita)

In particular, Wang wants to incorporate this technology Can detect manipulated images in BytePlus so that users can use it. This last mentioned platform is a subsidiary of ByteDance, which specializes in the development and research of artificial intelligence.

“For better or worse, deepfake is an emerging technology that simply cannot be ignored”Wang made the comments after winning the competition. According to Singapore TV’s CNA, Wan decided to participate because he has a strong interest in solving big challenges facing the media.

Continue reading


Source: Info Bae

Leave a Reply

Your email address will not be published. Required fields are marked *

Exit mobile version