Fusion-SSAT: Unleashing the Potential of Self-supervised Auxiliary Task by Feature Fusion for Generalized Deepfake Detection
This addresses the challenge of detecting deepfakes across diverse datasets, which is crucial for security and media integrity, but it appears incremental as it builds on existing self-supervised and fusion methods.
The paper tackled the problem of generalized deepfake detection by using self-supervised learning as an auxiliary task with feature fusion, achieving better cross-dataset generalizability compared to state-of-the-art detectors.
In this work, we attempted to unleash the potential of self-supervised learning as an auxiliary task that can optimise the primary task of generalised deepfake detection. To explore this, we examined different combinations of the training schemes for these tasks that can be most effective. Our findings reveal that fusing the feature representation from self-supervised auxiliary tasks is a powerful feature representation for the problem at hand. Such a representation can leverage the ultimate potential and bring in a unique representation of both the self-supervised and primary tasks, achieving better performance for the primary task. We experimented on a large set of datasets, which includes DF40, FaceForensics++, Celeb-DF, DFD, FaceShifter, UADFV, and our results showed better generalizability on cross-dataset evaluation when compared with current state-of-the-art detectors.