Ensemble-Based Deepfake Detection using State-of-the-Art Models with Robust Cross-Dataset Generalisation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wahab, Haroon, Ugail, Hassan, Jaleel, Lujain
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908440967774208
author Wahab, Haroon
Ugail, Hassan
Jaleel, Lujain
author_facet Wahab, Haroon
Ugail, Hassan
Jaleel, Lujain
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2507_05996
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Ensemble-Based Deepfake Detection using State-of-the-Art Models with Robust Cross-Dataset Generalisation
Wahab, Haroon
Ugail, Hassan
Jaleel, Lujain
Computer Vision and Pattern Recognition
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.
title Ensemble-Based Deepfake Detection using State-of-the-Art Models with Robust Cross-Dataset Generalisation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.05996