Online Detection of AI-Generated Images
This addresses the challenge of distinguishing AI-generated from real images for security and media integrity, but it is incremental as it builds on existing detection methods by adapting to streaming data and mixed-content scenarios.
The paper tackles the problem of detecting AI-generated images in a realistic online setting where new generators are continuously released, and images may contain both real and generated components. It demonstrates strong performance by training on historical models and testing on future ones, and extends this to pixel-level detection for inpainted images.
With advancements in AI-generated images coming on a continuous basis, it is increasingly difficult to distinguish traditionally-sourced images (e.g., photos, artwork) from AI-generated ones. Previous detection methods study the generalization from a single generator to another in isolation. However, in reality, new generators are released on a streaming basis. We study generalization in this setting, training on N models and testing on the next (N+k), following the historical release dates of well-known generation methods. Furthermore, images increasingly consist of both real and generated components, for example through image inpainting. Thus, we extend this approach to pixel prediction, demonstrating strong performance using automatically-generated inpainted data. In addition, for settings where commercial models are not publicly available for automatic data generation, we evaluate if pixel detectors can be trained solely on whole synthetic images.