Self-Supervised AI-Generated Image Detection: A Camera Metadata Perspective

TL;DR

This paper presents a self-supervised method for detecting AI-generated images by leveraging camera metadata, specifically EXIF tags, to learn intrinsic photographic features that generalize well across different generative models and real-world conditions.

Contribution

It introduces a novel self-supervised approach using EXIF metadata for AI-generated image detection, improving cross-model generalization and robustness over existing methods.

Findings

Significantly outperforms previous detectors in diverse scenarios.

Demonstrates strong generalization to in-the-wild samples.

Shows robustness to common image perturbations.

Abstract

The proliferation of AI-generated imagery poses escalating challenges for multimedia forensics, yet many existing detectors depend on assumptions about the internals of specific generative models, limiting their cross-model applicability. We introduce a self-supervised approach for detecting AI-generated images that leverages camera metadata -- specifically exchangeable image file format (EXIF) tags -- to learn features intrinsic to digital photography. Our pretext task trains a feature extractor solely on camera-captured photographs by classifying categorical EXIF tags (\emph{e.g.}, camera model and scene type) and pairwise-ranking ordinal and continuous EXIF tags (\emph{e.g.}, focal length and aperture value). Using these EXIF-induced features, we first perform one-class detection by modeling the distribution of photographic images with a Gaussian mixture model and flagging low-likelihood samples as AI-generated. We then extend to binary detection that treats the learned extractor as a strong regularizer for a classifier of the same architecture, operating on high-frequency residuals from spatially scrambled patches. Extensive experiments across various generative models demonstrate that our EXIF-induced detectors substantially advance the state of the art, delivering strong generalization to in-the-wild samples and robustness to common benign image perturbations. The code and model are publicly available at https://github.com/Ekko-zn/SDAIE.