נושא הפרוייקט

מספר פרוייקט מחלקה שמות סטודנטים אימייל שמות מנחים

מנגנון captcha שבעזרתו נשפר את היכולת לזהות deep-fakes במדיה(וידאו ואודיו)

A captcha mechanism with which we will improve the ability to identify deep-fakes in media (video and audio)

תקציר בעיברית תקציר באנגלית

Generative deep learning models are able to create realistic audio and video. This technology has been used to impersonate the faces and voices of individuals. These “deepfakes” are being used to spread misinformation, enable scams, perform fraud, and blackmail the innocent. The technology continues to advance and today attackers have the ability to generate deepfakes in real-time. This new capability poses a significant threat to society as attackers begin to exploit the technology in advances social engineering attacks. 

We propose Deepfake-CAPTCHA (D-CAPTCHA): a system for automatically detecting deepfake through challeng response analysis. Instead of passively observing call content, we actively interact with the caller by requesting that he or she to perform a task (the challenge, e.g., press on the cheek or turn around in audio calls, and moving the head to the side or occlusion the face in video calls). The task is easy for a human to perform but extremely hard for a deepfake model to recreate due to limitations in attack practicality and technology. When a deepfake tries to perform the task, the resulting content (the response) will be severely distorted - making it easier for an anomaly detector, classifier, or even the victim to detect.