In this article, you will get all the information regarding Intel introduces the “world’s first” real-time deepfake detector
dated: 2022-11-19 22:52:41 .
Intel has reportedly unveiled the world’s first real-time deepfake detector (Opens in a new window). FakeCatcher is said to have a 96% accuracy rate and works by analyzing blood flow in video pixels using innovative photoplethysmography (PPG (Opens in a new window)).
Ilke Demir, a senior research scientist at Intel Labs, developed the FakeCatcher detector in collaboration with Umur Ciftci from the State University of New York at Binghamton. The real-time detector uses Intel hardware and software and runs on a server and interfaces via a web platform.
FakeCatcher differs from most deep learning-based detectors in that it looks for authentic clues in real videos, rather than looking at raw data to detect signs of inauthenticity. His method is based on PPG, a method of measuring the amount of light that blood vessels in living tissue absorb or reflect. As our heart pumps blood, the veins change color and these signals are picked up by technology to determine if the video is fake or not.
Speaking to VentureBeat, Demir (Opens in new window) said FakeCatcher is unique because PPG signals “have never been applied to the problem of deep spoofing before.” The detector collects these signals from 32 locations on the face before algorithms translate them into spatio-temporal maps before making a decision on whether the video is real or fake.
Deepfake videos are a growing threat worldwide. According to Gartner (Opens in a new window), organizations will spend approximately $188 billion on cybersecurity solutions to counter them. Currently, detection applications typically require streaming video for analysis, and results can take hours.
Intel says the detector could be used by social media platforms to prevent users from uploading malicious deepfakes, while news organizations could use it to prevent the publication of accidentally manipulated videos.
Recommendation of our editors
Deepfakes target prominent politicians and celebrities. Last month, a virally altered TikTok (Opens in a new window) was designed to make it look like Joe Biden was singing the nursery rhyme Baby Shark instead of the national anthem.
Efforts to detect deepfakes have also run into issues of racial bias in the datasets used to train them. According to a 2021 University of Southern California study (Opens in a new window), some detectors showed a difference in error rate of up to 10.7% depending on racial group.
Do you like what you read?
Sign up for Security guard Newsletters for our top privacy and security stories, delivered straight to your inbox.
Intel introduces the “world’s first” real-time deepfake detector
Latest News by IlmHunt.com