Generalizable AI-Generated Image Detection Based on Fractal Self-Similarity in the Spectrum
This addresses the critical challenge of generalizable detection for AI-generated images, which is important for security and media integrity, though it appears incremental as it builds on existing spectral analysis approaches.
The paper tackles the generalization problem in AI-generated image detection by proposing a method based on fractal self-similarity in the spectrum, which improves detection accuracy for images from unseen generative models.
The generalization performance of AI-generated image detection remains a critical challenge. Although most existing methods perform well in detecting images from generative models included in the training set, their accuracy drops significantly when faced with images from unseen generators. To address this limitation, we propose a novel detection method based on the fractal self-similarity of the spectrum, a common feature among images generated by different models. Specifically, we demonstrate that AI-generated images exhibit fractal-like spectral growth through periodic extension and low-pass filtering. This observation motivates us to exploit the similarity among different fractal branches of the spectrum. Instead of directly analyzing the spectrum, our method mitigates the impact of varying spectral characteristics across different generators, improving detection performance for images from unseen models. Experiments on a public benchmark demonstrated the generalized detection performance across both GANs and diffusion models.