Generalizable AI-Generated Image Detection Based on Fractal Self-Similarity in the Spectrum
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910870119907328 |
|---|---|
| author | Xiao, Shengpeng Guo, Yuanfang Peng, Heqi Liu, Zeming Yang, Liang Wang, Yunhong |
| author_facet | Xiao, Shengpeng Guo, Yuanfang Peng, Heqi Liu, Zeming Yang, Liang Wang, Yunhong |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_08484 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Generalizable AI-Generated Image Detection Based on Fractal Self-Similarity in the Spectrum Xiao, Shengpeng Guo, Yuanfang Peng, Heqi Liu, Zeming Yang, Liang Wang, Yunhong Computer Vision and Pattern Recognition 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. |
| title | Generalizable AI-Generated Image Detection Based on Fractal Self-Similarity in the Spectrum |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.08484 |