Rethinking Cross-Generator Image Forgery Detection through DINOv3

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Zhenglin, Li, Jason, Wen, Haiquan, Li, Tianxiao, Yang, Xi, Qi, Lu, Peng, Bei, Huang, Xiaowei, Yang, Ming-Hsuan, Cheng, Guangliang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915642394804224
author Huang, Zhenglin
Li, Jason
Wen, Haiquan
Li, Tianxiao
Yang, Xi
Qi, Lu
Peng, Bei
Huang, Xiaowei
Yang, Ming-Hsuan
Cheng, Guangliang
author_facet Huang, Zhenglin
Li, Jason
Wen, Haiquan
Li, Tianxiao
Yang, Xi
Qi, Lu
Peng, Bei
Huang, Xiaowei
Yang, Ming-Hsuan
Cheng, Guangliang
contents As generative models become increasingly diverse and powerful, cross-generator detection has emerged as a new challenge. Existing detection methods often memorize artifacts of specific generative models rather than learning transferable cues, leading to substantial failures on unseen generators. Surprisingly, this work finds that frozen visual foundation models, especially DINOv3, already exhibit strong cross-generator detection capability without any fine-tuning. Through systematic studies on frequency, spatial, and token perspectives, we observe that DINOv3 tends to rely on global, low-frequency structures as weak but transferable authenticity cues instead of high-frequency, generator-specific artifacts. Motivated by this insight, we introduce a simple, training-free token-ranking strategy followed by a lightweight linear probe to select a small subset of authenticity-relevant tokens. This token subset consistently improves detection accuracy across all evaluated datasets. Our study provides empirical evidence and a feasible hypothesis for understanding why foundation models generalize across diverse generators, offering a universal, efficient, and interpretable baseline for image forgery detection.
format Preprint
id arxiv_https___arxiv_org_abs_2511_22471
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Cross-Generator Image Forgery Detection through DINOv3
Huang, Zhenglin
Li, Jason
Wen, Haiquan
Li, Tianxiao
Yang, Xi
Qi, Lu
Peng, Bei
Huang, Xiaowei
Yang, Ming-Hsuan
Cheng, Guangliang
Computer Vision and Pattern Recognition
As generative models become increasingly diverse and powerful, cross-generator detection has emerged as a new challenge. Existing detection methods often memorize artifacts of specific generative models rather than learning transferable cues, leading to substantial failures on unseen generators. Surprisingly, this work finds that frozen visual foundation models, especially DINOv3, already exhibit strong cross-generator detection capability without any fine-tuning. Through systematic studies on frequency, spatial, and token perspectives, we observe that DINOv3 tends to rely on global, low-frequency structures as weak but transferable authenticity cues instead of high-frequency, generator-specific artifacts. Motivated by this insight, we introduce a simple, training-free token-ranking strategy followed by a lightweight linear probe to select a small subset of authenticity-relevant tokens. This token subset consistently improves detection accuracy across all evaluated datasets. Our study provides empirical evidence and a feasible hypothesis for understanding why foundation models generalize across diverse generators, offering a universal, efficient, and interpretable baseline for image forgery detection.
title Rethinking Cross-Generator Image Forgery Detection through DINOv3
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.22471