A Dataless FaceSwap Detection Approach Using Synthetic Images

TL;DR

This paper introduces a dataless deepfake detection method that uses synthetic images generated by StyleGAN3, achieving comparable or better performance than traditional methods and reducing ethnicity bias.

Contribution

The authors propose a novel dataless face swap detection approach utilizing synthetic data, eliminating the need for real training data and improving generalization and fairness.

Findings

Performs on par with real-data trained models

Shows better generalization with limited real data

Reduces ethnicity bias in deepfake detection

Abstract

Face swapping technology used to create "Deepfakes" has advanced significantly over the past few years and now enables us to create realistic facial manipulations. Current deep learning algorithms to detect deepfakes have shown promising results, however, they require large amounts of training data, and as we show they are biased towards a particular ethnicity. We propose a deepfake detection methodology that eliminates the need for any real data by making use of synthetically generated data using StyleGAN3. This not only performs at par with the traditional training methodology of using real data but it shows better generalization capabilities when finetuned with a small amount of real data. Furthermore, this also reduces biases created by facial image datasets that might have sparse data from particular ethnicities.