Intel introduced the world’s first real-time deep fake detector. The developers claim Fake Catcher It has an accuracy of 96% and works by analyzing blood flow using innovative photoplethysmography (PPG).

İlke Demir, a senior scientist at Intel Labs, developed the FakeCatcher detector with Umur Çiftçi of the State University of New York in Binghamton. The detector uses Intel hardware and software, runs on a server and interacts via a web platform.

FakeCatcher differs from most deep learning-based detectors in that it looks for real clues in real videos rather than sifting through raw data for signs of inaccuracy. His method is based on PPG, a method used to measure the amount of light absorbed or reflected by blood vessels in living tissue. As our heart pumps blood, veins change color and these signals are collected by technology to determine if a video is fake.

FakeCatcher is unique because PPG signals “have not been applied to the deepfake issue before.” The detector collects these signals from 32 points on the face, transforms them into spatio-temporal maps before algorithms decide whether the video is real or fake. İlzi Demir

Deepfakes are a threat all over the world. According to Gartner, companies will spend approximately $188 billion on cybersecurity solutions against deep fraud. Currently, detection programs often require video uploads for analysis, and results can take several hours. Intel says the FakeCatcher detector can be used by social media platforms and news sources to avoid posting deep fakes. Source