Bridging the Gap: A Framework for Real-World Video Deepfake Detection via Social Network Compression Emulation

2025-08-12

AI Summary

Social media platforms’ proprietary video compression (e.g., YouTube, Facebook) erases low-level forensic traces, severely degrading the generalizability of deepfake detectors in real-world scenarios. To address this, we propose a few-shot-driven local compression simulation framework that faithfully reproduces multi-platform private compression artifacts—without requiring API access or real platform data. Our core contribution is the first joint mechanism integrating reversible parameter estimation and re-compression simulation: from a small set of uploaded videos, it accurately infers platform-specific compression parameters and synthesizes large-scale training data exhibiting authentic distortion characteristics. Evaluation on FaceForensics++ shows that detectors fine-tuned solely on our simulated data achieve detection performance on real uploaded videos comparable to models trained on actual shared platform data—effectively bridging the critical performance gap between controlled laboratory evaluation and practical deployment.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The growing presence of AI-generated videos on social networks poses new challenges for deepfake detection, as detectors trained under controlled conditions often fail to generalize to real-world scenarios. A key factor behind this gap is the aggressive, proprietary compression applied by platforms like YouTube and Facebook, which launder low-level forensic cues. However, replicating these transformations at scale is difficult due to API limitations and data-sharing constraints. For these reasons, we propose a first framework that emulates the video sharing pipelines of social networks by estimating compression and resizing parameters from a small set of uploaded videos. These parameters enable a local emulator capable of reproducing platform-specific artifacts on large datasets without direct API access. Experiments on FaceForensics++ videos shared via social networks demonstrate that our emulated data closely matches the degradation patterns of real uploads. Furthermore, detectors fine-tuned on emulated videos achieve comparable performance to those trained on actual shared media. Our approach offers a scalable and practical solution for bridging the gap between lab-based training and real-world deployment of deepfake detectors, particularly in the underexplored domain of compressed video content.

Problem

• Detecting deepfakes in real-world social network videos
• Overcoming platform compression challenges for detection
• Emulating social network video sharing pipelines

Innovation

• Emulates social network video compression
• Estimates parameters from uploaded videos
• Enables scalable deepfake detection training