HiMix: Hierarchical Artifact-aware Mixup for Generalized Synthetic Image Detection

TL;DR

HiMix is a novel framework that improves synthetic image detection by expanding training data and enhancing artifact-aware features, leading to better generalization to unseen generators.

Contribution

It introduces Mixup-driven Distributional Augmentation and Hierarchical Artifact-aware Representation modules for robust, generalized synthetic image detection.

Findings

Achieves state-of-the-art performance on multiple benchmarks.

Effectively improves generalization to unseen forgery methods.

Enhances sensitivity to low-level artifacts through novel augmentation.

Abstract

The rapid evolution of generative models has enabled the creation of highly realistic and diverse synthetic images, posing significant challenges to reliable and generalizable Synthetic Image Detection (SID). However, existing detectors are typically trained on limited and biased datasets, resulting in poor generalization to unseen generators. To address this issue, we propose HiMix, a unified framework that enhances generalization by expanding the training distribution and promoting artifact-aware representations. Specifically, the Mixup-driven Distributional Augmentation (MDA) module constructs continuous transitional samples between real and fake images, improving coverage of low-confidence regions and exposing the model to more challenging samples, while the pixel-wise mixup operation smoothly perturbs semantics to enhance sensitivity to low-level artifacts. Moreover, the Hierarchical Artifact-aware Representation (HAR) module aggregates artifact information from both global and local levels through cross-layer integration and coarse-to-fine feature fusion, enabling the extraction of discriminative forgery representations under diverse distributions. Extensive experiments across multiple benchmarks demonstrate that HiMix achieves state-of-the-art performance, establishing well-separated logits for improved generalization to unseen forgeries.