Researchers at the U.S. Army Combat Capabilities Development Command's Army Research Laboratory and the University of Southern California (USC) have developed a deepfake detection method for supporting mission-essential tasks.
The team said DefakeHop's core innovation is Successive Subspace Learning (SSL), a signal representation and transform theory designed as a neural network architecture.
USC's C.-C. Jay Kuo described SSL as "a complete data-driven unsupervised framework [that] offers a brand new tool for image processing and understanding tasks such as face biometrics."
Among DefakeHop's purported advantages over current state-of-the-art deepfake video detection methods are mathematical transparency, less complexity, and robustness against adversarial attacks.
From U.S. Army Research Laboratory
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA