Any-Resolution AI-Generated Image Detection by Spectral Learning

TL;DR

This paper introduces SPAI, a spectral learning-based method for detecting AI-generated images that generalizes across different models and resolutions, achieving state-of-the-art accuracy and robustness.

Contribution

The paper proposes a novel spectral learning framework with spectral reconstruction similarity and spectral context attention for generalizable AI-generated image detection.

Findings

Achieves 5.5% absolute AUC improvement over previous methods.

Demonstrates robustness against online perturbations.

Effective across 13 recent generative models.

Abstract

Recent works have established that AI models introduce spectral artifacts into generated images and propose approaches for learning to capture them using labeled data. However, the significant differences in such artifacts among different generative models hinder these approaches from generalizing to generators not seen during training. In this work, we build upon the key idea that the spectral distribution of real images constitutes both an invariant and highly discriminative pattern for AI-generated image detection. To model this under a self-supervised setup, we employ masked spectral learning using the pretext task of frequency reconstruction. Since generated images constitute out-of-distribution samples for this model, we propose spectral reconstruction similarity to capture this divergence. Moreover, we introduce spectral context attention, which enables our approach to efficiently capture subtle spectral inconsistencies in images of any resolution. Our spectral AI-generated image detection approach (SPAI) achieves a 5.5% absolute improvement in AUC over the previous state-of-the-art across 13 recent generative approaches, while exhibiting robustness against common online perturbations. Code is available on https://mever-team.github.io/spai.