Beyond Spectral Peaks: Interpreting the Cues Behind Synthetic Image Detection

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Mandelli, Sara, Vila-Portela, Diego, Vázquez-Padín, David, Bestagini, Paolo, Pérez-González, Fernando
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914079066554368
author Mandelli, Sara
Vila-Portela, Diego
Vázquez-Padín, David
Bestagini, Paolo
Pérez-González, Fernando
author_facet Mandelli, Sara
Vila-Portela, Diego
Vázquez-Padín, David
Bestagini, Paolo
Pérez-González, Fernando
contents Over the years, the forensics community has proposed several deep learning-based detectors to mitigate the risks of generative AI. Recently, frequency-domain artifacts (particularly periodic peaks in the magnitude spectrum), have received significant attention, as they have been often considered a strong indicator of synthetic image generation. However, state-of-the-art detectors are typically used as black-boxes, and it still remains unclear whether they truly rely on these peaks. This limits their interpretability and trust. In this work, we conduct a systematic study to address this question. We propose a strategy to remove spectral peaks from images and analyze the impact of this operation on several detectors. In addition, we introduce a simple linear detector that relies exclusively on frequency peaks, providing a fully interpretable baseline free from the confounding influence of deep learning. Our findings reveal that most detectors are not fundamentally dependent on spectral peaks, challenging a widespread assumption in the field and paving the way for more transparent and reliable forensic tools.
format Preprint
id arxiv_https___arxiv_org_abs_2510_05633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Spectral Peaks: Interpreting the Cues Behind Synthetic Image Detection
Mandelli, Sara
Vila-Portela, Diego
Vázquez-Padín, David
Bestagini, Paolo
Pérez-González, Fernando
Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
Over the years, the forensics community has proposed several deep learning-based detectors to mitigate the risks of generative AI. Recently, frequency-domain artifacts (particularly periodic peaks in the magnitude spectrum), have received significant attention, as they have been often considered a strong indicator of synthetic image generation. However, state-of-the-art detectors are typically used as black-boxes, and it still remains unclear whether they truly rely on these peaks. This limits their interpretability and trust. In this work, we conduct a systematic study to address this question. We propose a strategy to remove spectral peaks from images and analyze the impact of this operation on several detectors. In addition, we introduce a simple linear detector that relies exclusively on frequency peaks, providing a fully interpretable baseline free from the confounding influence of deep learning. Our findings reveal that most detectors are not fundamentally dependent on spectral peaks, challenging a widespread assumption in the field and paving the way for more transparent and reliable forensic tools.
title Beyond Spectral Peaks: Interpreting the Cues Behind Synthetic Image Detection
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2510.05633