FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learning

TL;DR

FReTAL is a transfer learning approach that employs knowledge distillation and representation learning to adapt deepfake detection models to new data types while preventing catastrophic forgetting, achieving high accuracy on diverse datasets.

Contribution

The paper introduces FReTAL, a novel transfer learning method combining knowledge distillation and representation learning for effective domain adaptation in deepfake detection.

Findings

FReTAL outperforms baseline methods in deepfake domain adaptation.

Achieves up to 86.97% accuracy on low-quality deepfakes.

Effectively prevents catastrophic forgetting during adaptation.

Abstract

As GAN-based video and image manipulation technologies become more sophisticated and easily accessible, there is an urgent need for effective deepfake detection technologies. Moreover, various deepfake generation techniques have emerged over the past few years. While many deepfake detection methods have been proposed, their performance suffers from new types of deepfake methods on which they are not sufficiently trained. To detect new types of deepfakes, the model should learn from additional data without losing its prior knowledge about deepfakes (catastrophic forgetting), especially when new deepfakes are significantly different. In this work, we employ the Representation Learning (ReL) and Knowledge Distillation (KD) paradigms to introduce a transfer learning-based Feature Representation Transfer Adaptation Learning (FReTAL) method. We use FReTAL to perform domain adaptation tasks on new deepfake datasets while minimizing catastrophic forgetting. Our student model can quickly adapt to new types of deepfake by distilling knowledge from a pre-trained teacher model and applying transfer learning without using source domain data during domain adaptation. Through experiments on FaceForensics++ datasets, we demonstrate that FReTAL outperforms all baselines on the domain adaptation task with up to 86.97% accuracy on low-quality deepfakes.