CLIP-Flow: A Universal Discriminator for AI-Generated Images Inspired by Anomaly Detection
This addresses security concerns by improving detection robustness for unseen models, though it is incremental as it builds on existing anomaly detection and CLIP methods.
The paper tackles the problem of detecting AI-generated images from unseen generative models by proposing a universal discriminator based on anomaly detection, achieving effective detection across various generators without needing access to AI-generated images during training.
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.