Supervised Contrastive Learning for Few-Shot AI-Generated Image Detection and Attribution
This addresses the problem of digital media integrity for forensic systems, offering a scalable solution that adapts to new generative models without exhaustive retraining, though it builds incrementally on existing contrastive and few-shot learning methods.
The paper tackles the challenge of detecting AI-generated images and attributing them to specific generators, especially as new models emerge rapidly. It proposes a two-stage framework using supervised contrastive learning and few-shot learning, achieving 91.3% detection accuracy with only 150 images per class and significant improvements in attribution metrics.
The rapid advancement of generative artificial intelligence has enabled the creation of synthetic images that are increasingly indistinguishable from authentic content, posing significant challenges for digital media integrity. This problem is compounded by the accelerated release cycle of novel generative models, which renders traditional detection approaches (reliant on periodic retraining) computationally infeasible and operationally impractical. This work proposes a novel two-stage detection framework designed to address the generalization challenge inherent in synthetic image detection. The first stage employs a vision deep learning model trained via supervised contrastive learning to extract discriminative embeddings from input imagery. Critically, this model was trained on a strategically partitioned subset of available generators, with specific architectures withheld from training to rigorously ablate cross-generator generalization capabilities. The second stage utilizes a k-nearest neighbors (k-NN) classifier operating on the learned embedding space, trained in a few-shot learning paradigm incorporating limited samples from previously unseen test generators. With merely 150 images per class in the few-shot learning regime, which are easily obtainable from current generation models, the proposed framework achieves an average detection accuracy of 91.3\%, representing a 5.2 percentage point improvement over existing approaches . For the source attribution task, the proposed approach obtains improvements of of 14.70\% and 4.27\% in AUC and OSCR respectively on an open set classification context, marking a significant advancement toward robust, scalable forensic attribution systems capable of adapting to the evolving generative AI landscape without requiring exhaustive retraining protocols.