Logo image
Back
Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images
Conference proceeding

Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images

Tai D. Nguyen, Aref Azizpour and Matthew C. Stamm
Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online), pp 3040-3050
10 Jun 2025
DOI: https://doi.org/10.1109/CVPR52734.2025.00289

Abstract

Adaptation models ai generated image detection ai security Filters Forensics Generators Image coding Media Microstructure openset synthetic image attribution Resilience synthetic image attribution synthetic image detection Training unsupervised synthetic image source clustering Computer Vision Pattern Recognition
The emergence of advanced AI-based tools to generate realistic images poses significant challenges for forensic detection and source attribution, especially as new generative techniques appear rapidly. Traditional methods often fail to generalize to unseen generators due to reliance on features specific to known sources during training. To address this problem, we propose a novel approach that explicitly models forensic microstructures-subtle, pixel-level patterns unique to the image creation process. Using only real images in a self-supervised manner, we learn a set of diverse predictive filters to extract residuals that capture different aspects of these microstructures. By jointly modeling these residuals across multiple scales, we obtain a compact model whose parameters constitute a unique forensic self-description for each image. This self-description enables us to perform zero-shot detection of synthetic images, open-set source attribution of images, and clustering based on source without prior knowledge. Extensive experiments demonstrate that our method achieves superior accuracy and adaptability compared to competing techniques, advancing the state of the art in synthetic media forensics.

Metrics

1 Record Views
1 citations in Scopus

Related content

Has preprint

Details

Logo image