Leveraging Natural Frequency Deviation for Diffusion-Generated Image Detection

24 Sept 2024 (modified: 15 Nov 2024)ICLR 2025 Conference Withdrawn SubmissionEveryoneRevisionsBibTeXCC BY 4.0
Keywords: frequency domain, filter banks, diffusion-generated image detection
Abstract:

Diffusion models have achieved remarkable success in image synthesis, but the generated high-quality images raise concerns about potential malicious use. Existing detectors often struggle to capture distinctive features across different training models, limiting their generalization to unseen diffusion models with varying schedulers and hyperparameters. To address this issue, we observe that diffusion-generated images exhibit progressively larger differences from real images across low- to high-frequency bands. Based on this insight, we propose a novel image representation called \textbf{N}atural \textbf{F}r\textbf{e}quency \textbf{De}viation~(\textbf{DEFEND}). DEFEND applies a weighted filter to the Fourier spectrum, suppressing less discriminative bands while enhancing more informative ones. This approach, grounded in a comprehensive analysis of frequency-based differences between real and diffusion-generated images, enables robust detection of images from unseen diffusion models and provides resilience to various perturbations. Extensive experiments on diffusion-generated image datasets show that our method outperforms state-of-the-art detectors with superior generalization and robustness.

Supplementary Material: pdf
Primary Area: alignment, fairness, safety, privacy, and societal considerations
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 3874
Loading

OpenReview is a long-term project to advance science through improved peer review with legal nonprofit status. We gratefully acknowledge the support of the OpenReview Sponsors. © 2025 OpenReview