Through the Lens: Benchmarking Deepfake Detectors Against Moiré-Induced Distortions

2025-10-27

AI Summary

Deepfake detection faces a robustness bottleneck in real-world scenarios due to Moiré interference—artifacts arising when smartphones capture digital displays—which severely degrades the performance of existing detectors. Method: We systematically evaluate 15 state-of-the-art (SOTA) deepfake detectors under Moiré distortion and introduce DMF, the first dedicated Deepfake Moiré video dataset, comprising both real-world captures and controllably synthesized data. We further analyze the impact of conventional Moiré removal preprocessing. Contribution/Results: Moiré distortion reduces detector accuracy by up to 25.4%; critically, standard denoising and demoiréing pre-processing further degrades accuracy by 17.2%, revealing fundamental limitations of traditional pipeline-based approaches in complex imaging chains. Our findings underscore the necessity of end-to-end robust modeling and establish DMF as the first benchmark and reproducible analytical framework for real-world deepfake detection under Moiré interference.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Mixture of Experts (MoE)
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Deepfake detection remains a pressing challenge, particularly in real-world settings where smartphone-captured media from digital screens often introduces Moiré artifacts that can distort detection outcomes. This study systematically evaluates state-of-the-art (SOTA) deepfake detectors on Moiré-affected videos, an issue that has received little attention. We collected a dataset of 12,832 videos, spanning 35.64 hours, from the Celeb-DF, DFD, DFDC, UADFV, and FF++ datasets, capturing footage under diverse real-world conditions, including varying screens, smartphones, lighting setups, and camera angles. To further examine the influence of Moiré patterns on deepfake detection, we conducted additional experiments using our DeepMoiréFake, referred to as (DMF) dataset and two synthetic Moiré generation techniques. Across 15 top-performing detectors, our results show that Moiré artifacts degrade performance by as much as 25.4%, while synthetically generated Moiré patterns lead to a 21.4% drop in accuracy. Surprisingly, demoiréing methods, intended as a mitigation approach, instead worsened the problem, reducing accuracy by up to 17.2%. These findings underscore the urgent need for detection models that can robustly handle Moiré distortions alongside other realworld challenges, such as compression, sharpening, and blurring. By introducing the DMF dataset, we aim to drive future research toward closing the gap between controlled experiments and practical deepfake detection.

Problem

• Evaluating deepfake detectors on Moiré-affected real-world videos
• Assessing performance degradation from Moiré artifacts in detection
• Investigating mitigation approaches for Moiré distortions in deepfake detection

Innovation

• Systematically evaluates detectors on Moiré-affected videos
• Introduces DeepMoiréFake dataset for real-world distortions
• Shows demoiréing methods worsen deepfake detection accuracy