G.L. Chen and C.C. Hsu*, “Jointly Defending DeepFake Manipulation and Adversarial Attack using Decoy Mechanism,” IEEE T-PAMI （IF: 24.31), to appear.

This paper introduces a new method for joint deepfake image detection and an adversarial defense approach. We have created a baiting attack to deceive the “attacker” and mislead the direction of the attack, making the counter-attack ineffective, thus achieving excellent and stable performance in deepfake image detection. We will soon share relevant data and code.

#tpami #t_pami
#acvlab #ncku
#deepfake
#adversarial_defense
#decoy
#security