Deepfakes, particularly those involving faceswap-based manipulations, have sparked significant societal concern due to their increasing realism and potential for misuse. Despite rapid advancements in generative models, detection methods have not kept pace, creating a critical gap in defense strategies.
Abstract
This disparity is further amplified by the disconnect between academic research and real-world applications, which often prioritize different objectives and evaluation criteria. In this study, we take a pivotal step toward bridging this gap by presenting a novel observation: the post-processing step of super-resolution, commonly employed in real-world scenarios, substantially undermines the effectiveness of existing deepfake detection methods.
Novel Contributions
Real-World Dataset
To substantiate this claim, we introduce and publish the first real-world faceswap dataset, collected from popular online faceswap platforms.
Performance Evaluation
We then qualitatively evaluate the performance of state-of-the-art deepfake detectors on real-world deepfakes, revealing that their accuracy approaches the level of random guessing.
Quantitative Analysis
Furthermore, we quantitatively demonstrate the significant performance degradation caused by common post-processing techniques.
Key Findings
Critical Gap Identified
Academic vs Reality: Significant disconnect between academic benchmarks and real-world performance
Post-processing Impact: Super-resolution and other common post-processing steps severely degrade detection accuracy
Random Performance: State-of-the-art detectors perform close to random guessing on real-world data
Real-World Challenges
Platform Variations: Different online platforms employ various post-processing techniques
Quality Enhancement: Super-resolution improves visual quality but breaks detection methods
Evaluation Gaps: Current academic evaluation doesn't reflect real-world usage
Practical Impact
By addressing this overlooked challenge, our study underscores a critical avenue for enhancing the robustness and practical applicability of deepfake detection methods in real-world settings.
Subjects
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Citation
arXiv:2502.10920 [cs.CV] (or arXiv:2502.10920v1 [cs.CV] for this version)
