用异常检测思路打造通用AI图像检测器，无需见过的生成图也能识别。
CLIP-Flow: A Universal Discriminator for AI-Generated Images Inspired by Anomaly Detection
- 基于异常检测，用自然图的频域扰动图作代理训练，不依赖真实生成图。
- 在多个未见过的生成模型上检测准确率超90%，泛化能力显著。
- 适合需要应对未知生成模型的AI内容安全场景，如平台审核。
随着AI生成模型的快速发展，生成图像（AIIs）的视觉质量已接近真实图像，引发安全担忧。现有检测方法多采用传统分类范式，以自然图像和已知生成模型的AI图为训练数据，导致对未见生成模型的检测性能有限。为此，本文提出一种基于异常检测的通用AI图像检测框架。该判别器无需访问任何真实的生成图像，通过无监督学习构建可泛化的表征。具体而言，利用预训练的CLIP编码器提取特征，并设计类似归一化流的无监督模型。训练时使用代理图像（如对自然图像施加频域修改操作得到的图像），通过最小化代理图像的似然概率进行优化，可选地结合最大化自然图像的似然。大量实验表明，该方法在多种不同生成模型产生的AI图像上均表现优异，具备强泛化能力。
原文摘要 · Abstract (English)
With the rapid advancement of AI generative models, the visual quality of AI-generated images (AIIs) has become increasingly close to natural images, which inevitably raises security concerns. Most AII detectors often employ the conventional image classification pipeline with natural images and AIIs (generated by a generative model), which can result in limited detection performance for AIIs from unseen generative models. To solve this, we proposed a universal AI-generated image detector from the perspective of anomaly detection. Our discriminator does not need to access any AIIs and learn a generalizable representation with unsupervised learning. Specifically, we use the pre-trained CLIP encoder as the feature extractor and design a normalizing flow-like unsupervised model. Instead of AIIs, proxy images, e.g., obtained by applying a spectral modification operation on natural images, are used for training. Our models are trained by minimizing the likelihood of proxy images, optionally combined with maximizing the likelihood of natural images. Extensive experiments demonstrate the effectiveness of our method on AIIs produced by various image generators.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。