Bilkent University
Department of Computer Engineering
CS 590/690 SEMINAR
Generalizable Face Forgery Detection via Triplet Learning
Mustafa Hakan Kara
Master Student
(Supervisor: Prof.Dr.Uğur Güdükbay)
Computer Engineering Department
Bilkent University
Abstract: As AI-generated face forgeries, such as deepfakes, become increasingly sophisticated, the challenge of detecting these manipulations grows, underscoring the importance of maintaining digital media integrity and combating visual disinformation. Current detection models, predominantly based on supervised training with domain-specific data, may falter against face forgeries generated from unencountered techniques. This research aims to develop a universal face forgery detection framework that employs triplet learning in Siamese networks for enhanced adaptability across diverse forgery methods. The framework uses triplet learning to capture fine-grained nuances in facial features, diminishing reliance on domain-specific training and improving the robustness of digital media verification against a broad spectrum of face forgeries.
DATE: March 18, Monday @ 14:30 Place: EA 502