Facebook AI can reverse engineer deepfakes and figure out their source, but the Intel FakeCatcher is the world’s first real-time deepfake detector. How does it work? This technology runs on Intel hardware and software, both of which interface through a web-based platform. Optimized FakeCatcher architecture consists of teams using OpenVino to run AI models for face and landmark detection algorithms.
Unlike other deep learning-based detectors that look at raw data to try to find signs of in-authenticity, FakeCatcher seeks authentic clues in real videos by simply assessing what makes us human, such as subtle “blood flow” in the pixels of a video. Simply put, when our hearts pump blood, the veins change color. These signals are collected from all over the face and algorithms then translate these signals into spatiotemporal maps for the deep learning algorithms that can instantly detect whether a video is real or fake.
No products found.
Deepfake videos are everywhere now. You have probably already seen them; videos of celebrities doing or saying things they never actually did,” said Ilke Demir, senior staff research scientist in Intel Labs.
