Ensemble-Based Deepfake Detection using State-of-the-Art Models with Robust Cross-Dataset Generalisation
This provides a robust solution for real-world deepfake detection, where forgery types are unknown, though it is incremental as it builds on existing models.
The paper tackled the problem of deepfake detection models performing poorly on out-of-distribution data by using an ensemble-based approach with state-of-the-art models, resulting in more stable and reliable performance across diverse datasets.
Machine learning-based Deepfake detection models have achieved impressive results on benchmark datasets, yet their performance often deteriorates significantly when evaluated on out-of-distribution data. In this work, we investigate an ensemble-based approach for improving the generalization of deepfake detection systems across diverse datasets. Building on a recent open-source benchmark, we combine prediction probabilities from several state-of-the-art asymmetric models proposed at top venues. Our experiments span two distinct out-of-domain datasets and demonstrate that no single model consistently outperforms others across settings. In contrast, ensemble-based predictions provide more stable and reliable performance in all scenarios. Our results suggest that asymmetric ensembling offers a robust and scalable solution for real-world deepfake detection where prior knowledge of forgery type or quality is often unavailable.