MagDR: Mask-guided Detection and Reconstruction for Defending Deepfakes

Abstract

Deepfakes raised serious concerns on the authenticity of visual contents. Prior works revealed the possibility to disrupt deepfakes by adding adversarial perturbations to the source data, but we argue that the threat has not been eliminated yet. This paper presents MagDR, a mask-guided detection and reconstruction pipeline for defending deepfakes from adversarial attacks. MagDR starts with a detection module that defines a few criteria to judge the abnormality of the output of deepfakes, and then uses it to guide a learnable reconstruction procedure. Adaptive masks are extracted to capture the change in local facial regions. In experiments, MagDR defends three main tasks of deepfakes, and the learned reconstruction pipeline transfers across input data, showing promising performance in defending both black-box and white-box attacks.

arXiv commentAccepted to CVPR2021

References in corpus (11)

• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Towards Deep Learning Models Resistant to Adversarial Attacks
• On Evaluating Adversarial Robustness
• A study of the effect of JPG compression on adversarial images
• FaceShifter: Towards High Fidelity And Occlusion Aware Face Swapping
• Facial Soft Biometrics for Recognition in the Wild: Recent Works, Annotation, and COTS Evaluation
• MaskGAN: Towards Diverse and Interactive Facial Image Manipulation
• ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation
• Appending Adversarial Frames for Universal Video Attack