Deepkfake is a real plague on the internet and especially on YouTube, where it is sometimes difficult to tell the real from the fake. But Intel seems to have the solution.
Intel therefore presents a technological solution called FakeCatche, which aims to analyze a supposed deepfake in order to unravel the true from the false. For those who don’t know, a deepfake is a technology that mimics the human face on another face. This is what some film studios use (in larger proportions) to rejuvenate an actor’s face.
Intel analyzes deepfakes
Intel therefore offers FakeCatcher, which focuses on physical details for its analysis. intel To explain especially :
Deepfake videos are everywhere now. You’ve probably seen them before; Videos of celebrities doing or saying things they’ve never done before.
FakeCatcher (Intel) looks for authentic clues in real videos and evaluates what makes us human
. A subtle “blood flow” in the pixels of a video. When our heart pumps blood, our veins change color. These blood flow signals are collected across the face and algorithms translate these signals into spatio-temporal maps. Then, thanks to deep learning, we can immediately see if a video is real or fake.
This almost instantaneous processing of video by Intel is a first in this field. Until now, it has been impossible to analyze them in real time. The manufacturer plans to offer this solution on social networks, the media and NGOs for large-scale deployment. A tool that could be very useful against fake news and generally against the transmission of false information about false personalities.
What do you think of this kind of technology?
