Active Fake: DeepFake Camouflage

View on arXiv ← Back to list

Authors: Pu Sun, Honggang Qi, Yuezun Li

Published: 2024-09-05 02:46:36+00:00

AI Summary

This paper introduces DeepFake Camouflage, a method to evade deepfake detectors by intentionally introducing blending inconsistencies in authentic videos. A novel framework, Camouflage GAN (CamGAN), is proposed to generate these imperceptible yet effective inconsistencies using adversarial learning and a reinforcement learning-based optimization.

Abstract

DeepFake technology has gained significant attention due to its ability to manipulate facial attributes with high realism, raising serious societal concerns. Face-Swap DeepFake is the most harmful among these techniques, which fabricates behaviors by swapping original faces with synthesized ones. Existing forensic methods, primarily based on Deep Neural Networks (DNNs), effectively expose these manipulations and have become important authenticity indicators. However, these methods mainly concentrate on capturing the blending inconsistency in DeepFake faces, raising a new security issue, termed Active Fake, emerges when individuals intentionally create blending inconsistency in their authentic videos to evade responsibility. This tactic is called DeepFake Camouflage. To achieve this, we introduce a new framework for creating DeepFake camouflage that generates blending inconsistencies while ensuring imperceptibility, effectiveness, and transferability. This framework, optimized via an adversarial learning strategy, crafts imperceptible yet effective inconsistencies to mislead forensic detectors. Extensive experiments demonstrate the effectiveness and robustness of our method, highlighting the need for further research in active fake detection.

Key findings

Extensive experiments demonstrate that CamGAN effectively deceives multiple deepfake detectors across white-box and black-box settings, outperforming existing adversarial attack methods. The generated inconsistencies are imperceptible to the human eye, highlighting a new challenge in active deepfake detection.

Approach

CamGAN uses adversarial learning to generate parameters for Gaussian noise and filtering operations applied to real faces. These operations introduce blending inconsistencies that mislead deepfake detectors while maintaining visual realism. Reinforcement learning addresses the non-differentiability of the image processing steps.

Datasets

FaceForensics++, Celeb-DF

Model(s)

Xception, FFD, SPSL, SRM

Author countries

China

← Previous