Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Learning

TL;DR

This paper introduces FreqNet, a frequency-aware deepfake detection method that enhances generalizability by focusing on high-frequency features and source-agnostic frequency domain learning, outperforming existing models across multiple GANs.

Contribution

The paper presents a novel frequency domain learning approach, FreqNet, which improves deepfake detection generalizability by emphasizing high-frequency information and source-agnostic features.

Findings

Achieves +9.8% accuracy over state-of-the-art methods.

Requires fewer parameters than existing models.

Demonstrates robustness across 17 different GANs.

Abstract

This research addresses the challenge of developing a universal deepfake detector that can effectively identify unseen deepfake images despite limited training data. Existing frequency-based paradigms have relied on frequency-level artifacts introduced during the up-sampling in GAN pipelines to detect forgeries. However, the rapid advancements in synthesis technology have led to specific artifacts for each generation model. Consequently, these detectors have exhibited a lack of proficiency in learning the frequency domain and tend to overfit to the artifacts present in the training data, leading to suboptimal performance on unseen sources. To address this issue, we introduce a novel frequency-aware approach called FreqNet, centered around frequency domain learning, specifically designed to enhance the generalizability of deepfake detectors. Our method forces the detector to continuously focus on high-frequency information, exploiting high-frequency representation of features across spatial and channel dimensions. Additionally, we incorporate a straightforward frequency domain learning module to learn source-agnostic features. It involves convolutional layers applied to both the phase spectrum and amplitude spectrum between the Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (iFFT). Extensive experimentation involving 17 GANs demonstrates the effectiveness of our proposed method, showcasing state-of-the-art performance (+9.8\%) while requiring fewer parameters. The code is available at {\cred \url{https://github.com/chuangchuangtan/FreqNet-DeepfakeDetection}}.