Exploring the Impact of Moire Pattern on Deepfake Detectors

View on arXiv ← Back to list

Authors: Razaib Tariq, Shahroz Tariq, Simon S. Woo

Published: 2024-07-15 02:39:24+00:00

AI Summary

This research investigates the impact of Moiré patterns, artifacts introduced when capturing videos from screens, on the accuracy of deepfake detectors. Experiments using CelebDF and FF++ datasets show a significant drop in detector accuracy, with none exceeding 68% on average, highlighting the need for improved real-world deepfake detection.

Abstract

Deepfake detection is critical in mitigating the societal threats posed by manipulated videos. While various algorithms have been developed for this purpose, challenges arise when detectors operate externally, such as on smartphones, when users take a photo of deepfake images and upload on the Internet. One significant challenge in such scenarios is the presence of Moir'e patterns, which degrade image quality and confound conventional classification algorithms, including deep neural networks (DNNs). The impact of Moir'e patterns remains largely unexplored for deepfake detectors. In this study, we investigate how camera-captured deepfake videos from digital screens affect detector performance. We conducted experiments using two prominent datasets, CelebDF and FF++, comparing the performance of four state-of-the-art detectors on camera-captured deepfake videos with introduced Moir'e patterns. Our findings reveal a significant decline in detector accuracy, with none achieving above 68% on average. This underscores the critical need to address Moir'e pattern challenges in real-world deepfake detection scenarios.

Key findings

The presence of Moiré patterns significantly reduces the accuracy of state-of-the-art deepfake detectors; none achieved above 68% accuracy on average. This underscores the vulnerability of current detectors to real-world capture conditions and the need for more robust methods.

Approach

The researchers captured deepfake videos from a computer screen using a smartphone camera to introduce Moiré patterns. They then evaluated the performance of four state-of-the-art deepfake detectors on these camera-captured videos and compared the results to their performance on original videos.

Datasets

CelebDF and FaceForensics++ (FF++)

Model(s)

CoReD, CLRNet, Xception, BZNet, QAD, and ADD

Author countries

South Korea, Australia

← Previous