# DFD-Lab: A Modular Audio-Visual Deepfake Detection Pipeline

arXiv: https://arxiv.org/abs/2609.23830 · HTML: https://mlsanity.com/paper/2609.23830/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 11 references that are in the map.
- First posted on arXiv 20 Sep 2026.

## Builds on (most cited first, 8 of 8)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [Deep Audio-Visual Speech Recognition](https://mlsanity.com/paper/1809.02108/) (2018)
- [DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection](https://mlsanity.com/paper/2001.03024/) (2020)
- [DeepfakeBench: A Comprehensive Benchmark of Deepfake Detection](https://mlsanity.com/paper/2307.01426/) (2023)
- [AVFF: Audio-Visual Feature Fusion for Video Deepfake Detection](https://mlsanity.com/paper/2406.02951/) (2024)
- [Deepfake-Eval-2024: A Multi-Modal In-the-Wild Benchmark of Deepfakes Circulated in 2024](https://mlsanity.com/paper/2503.02857/) (2025)
- [DeepfakeBench-MM: A Comprehensive Benchmark for Multimodal Deepfake Detection](https://mlsanity.com/paper/2510.22622/) (2025)

## Abstract (arXiv)

Comparing audio-visual deepfake detectors requires coordinating dataset adaptation, temporal input representation, model interfaces and experimental conditions. We present DFD-Lab, a modular pipeline that separates these responsibilities while supporting shared training and evaluation workflows. We integrate three implementations: Xception-based maximum-logit fusion, ResNet with temporal LSTM fusion, and our AVFF reimplementation. Experiments cover external testing, degradation-based training augmentation and evaluation-time corruption. On a filtered subset of Deepfake-Eval-2024, models trained on FakeAVCeleb attain baseline AUROC values of 0.504, 0.538 and 0.458. JPEG50 training augmentation raises these to 0.691, 0.605 and 0.570, respectively, while all three accuracies decrease. These results illustrate why training interventions, evaluation corruptions and metric-dependent outcomes should remain distinct within a common pipeline. The contribution is the integration of audio-visual processing, interchangeable detectors and configurable experimental workflows, supported by empirical case studies. The findings highlight the challenge of cross-dataset detection and the complementary information provided by ranking and classification metrics.