PGC: Peak-Guided Calibration for Generalizable AI-Generated Image Detection

TL;DR

The paper introduces Peak-Guided Calibration (PGC), a novel method that enhances AI-generated image detection by focusing on salient local features, improving accuracy across diverse datasets and real-world scenarios.

Contribution

It proposes a peak-sensitive aggregation strategy for better detection of subtle generative artifacts and introduces the challenging CommGen15 dataset for benchmarking.

Findings

PGC improves mean accuracy by +12.3% on CommGen15.

Achieves new state-of-the-art results on GenImage, AIGI, and UniversalFakeDetect.

Code is publicly available at https://github.com/xiaoyu6868/PGC.

Abstract

The rapid evolution of generative AI, from GANs to modern diffusion models, has resulted in increasingly subtle discriminative clues. These fine-grained signals are often overshadowed by dominant, high-fidelity image content (e.g., the main subject), limiting the reliability of existing detectors that predominantly rely on global representations. To address this challenge, we propose the Peak-Guided Calibration (PGC) framework. PGC introduces a novel strategy that aggregates salient features via a peak-focusing mechanism. Specifically, by employing a peak-sensitive aggregation that accentuates the most discriminative local clues, PGC leverages these critical signals to calibrate the global decision. This approach recovers subtle patterns that would otherwise be submerged in the global context. Furthermore, to better simulate real-world threats, we introduce the CommGen15 dataset, a challenging benchmark comprising samples from 15 commercial models. Extensive experiments demonstrate that PGC achieves state-of-the-art performance. Specifically, it improves mean accuracy by +12.3% on our CommGen15 dataset, and sets new records on standard benchmarks, including GenImage (+2.1%), AIGI (+3.5%), and UniversalFakeDetect (+1.7%). Code is available at https://github.com/xiaoyu6868/PGC.