AV-Deepfake1M++: A Large-Scale Audio-Visual Deepfake Benchmark with Real-World Perturbations

TL;DR

This paper introduces AV-Deepfake1M++, a large-scale dataset with 2 million videos featuring diverse deepfake manipulations and real-world perturbations, aimed at advancing deepfake detection research.

Contribution

It provides a comprehensive, diversified dataset with generation strategies and benchmarking tools, facilitating progress in deepfake detection and analysis.

Findings

Benchmarking results with state-of-the-art methods

Dataset diversity improves detection robustness

Launch of 2025 Deepfakes Detection Challenge

Abstract

The rapid surge of text-to-speech and face-voice reenactment models makes video fabrication easier and highly realistic. To encounter this problem, we require datasets that rich in type of generation methods and perturbation strategy which is usually common for online videos. To this end, we propose AV-Deepfake1M++, an extension of the AV-Deepfake1M having 2 million video clips with diversified manipulation strategy and audio-visual perturbation. This paper includes the description of data generation strategies along with benchmarking of AV-Deepfake1M++ using state-of-the-art methods. We believe that this dataset will play a pivotal role in facilitating research in Deepfake domain. Based on this dataset, we host the 2025 1M-Deepfakes Detection Challenge. The challenge details, dataset and evaluation scripts are available online under a research-only license at https://deepfakes1m.github.io/2025.