通过相位谱鲁棒性检测压缩后的AI生成图像
Detecting Compressed AI-Generated Images via Phase Spectrum Robustness
利用压缩下仍稳定的相位谱作为跨模态线索，并通过多域调制适配器实现参数高效微调，为视觉取证类任务提供了一条不依赖原图-压缩图配对与压缩标签的鲁棒特征融合思路。
原文依据 the phase spectrum remains stable under compression
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文旨在提出一个鲁棒的AI生成图像检测框架，以应对在线社交网络中图像压缩所导致的性能退化问题。关键挑战有两方面：1）压缩破坏了现有方法所依赖的脆弱伪影；2）压缩引入了新的压缩伪影，对检测造成干扰。现有方法通常通过收集原始图像-压缩图像对以及压缩标签来增强压缩鲁棒性，然而收集与标注过程资源消耗极大。为解决这些问题，我们提出压缩鲁棒的相位协调Transformer，其动机在于观察到相位谱在压缩下保持稳定。该框架包含一个相位协调跨模态交互模块，利用相位谱信息进行特征融合以增强压缩鲁棒性，以及一个多域调制适配器，在进一步精炼融合特征的同时实现参数高效微调。特别地，该框架无需压缩-原始数据对与压缩标签即可运行。在仅有少量压缩标签可用时，我们引入难度感知一致性损失，通过在训练中优先关注困难压缩样本以最大化其效用，从而进一步提升鲁棒性。大量实验表明，我们的方法显著优于最优方法，对图像压缩展现出更优的鲁棒性。
英文原文摘要 1514 字符
This paper aims to present a robust AI-generated image detection framework designed to address performance degradation caused by image compression in online social networks. The key challenges are twofold: 1) compression destroys fragile artifacts that are crucial to existing methods, and 2) it introduces new compression artifacts that interfere with detection. Existing methods typically enhance the compression robustness by collecting original-compression pairs and compression labels. However, the collection and annotation process is highly resource-intensive. To address these issues, we propose a Compression-Robust Phase-Harmonized Transformer, motivated by the observation that phase spectrum remains stable under compression. The framework consists of a phase-harmonized cross-modal interaction module that leverages phase spectrum information for feature fusion, enhancing compression robustness, and a multi-domain modulation adapter that further refines fused features while enabling parameter-efficient fine-tuning. In particular, the framework operates without requiring compression-original data pairs and compression labels. When limited compression labels are available, we introduce a difficulty-aware consistency loss to maximize their utility by prioritizing hard compressed samples during training, further boosting robustness. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches, exhibiting superior robustness against image compression.