Friday, December 9

Intel’s FakeCatcher can detect fake videos with 96% accuracy

Photo : IANS

Amid growing threat from misinformation, Intel has unveiled a new deepfake detection platform that can flag discrepancies with a 96% accuracy rate. The platform is the world’s first real-time deepfake detector that returns results in milliseconds..

What are deepfakes?

Deepfakes – a combination of the words ‘deep learning’ and ‘fake’ – are a kind of visual media wherein a person’s existing image or video is replaced with another individual’s likeness. The technique makes use of machine learning and artificial intelligence to create deceptive (yet realistic) content. In recent years, deepfakes have garnered focus for their potential use in pornographic material, fake news financial fraud and more.

How does FakeCatcher work?

According to the company, the Intel technology looks for authentic clues in real videos, by assessing subtle ‘blood flow’ in the pixels of a video.

“When our hearts pump blood, our veins change color. These blood flow signals are collected from all over the face and algorithms translate these signals into spatiotemporal maps. Then, using deep learning, we can instantly detect whether a video is real or fake,” the company said in a press release earlier this month.

How can the technology be utilised?

There are several potential uses for FakeCatcher. As the Intel press note reminds, social media platforms could leverage the technology to prevent users from uploading harmful deepfake videos. Global news organisations, meanwhile, could use the detector to avoid inadvertently amplifying manipulated videos.



Source link