RIGID: A Training-Free and Model-Agnostic Framework for Robust AI-Generated Image Detection

TMLR Paper6729 Authors

Abstract: The rapid advances in generative AI models have empowered the creation of highly realistic images with arbitrary content, raising concerns about potential misuse and harm, such as Deepfakes. Current research focuses on training detectors using large datasets of generated images. However, these training-based solutions are often computationally expensive and show limited generalization to unseen generated images. In this paper, we propose a training-free method to distinguish between real and AI-generated images. We first observe that real images are more robust to tiny noise perturbations than AI-generated images in the representation space of vision foundation models. Based on this observation, we propose RIGID, a training-free and model-agnostic method for robust AI-generated image detection. RIGID is a simple yet effective approach that identifies whether an image is AI-generated by comparing the representation similarity between the original and the noise-perturbed counterpart. Our comprehensive evaluation demonstrates RIGID’s exceptional performance. RIGID surpasses existing training-free detectors by more than 25% on average. Remarkably, RIGID performs comparably to training-based methods, particularly on unseen domain data. Additionally, RIGID maintains consistent performance across various image generation techniques and demonstrates strong resilience to common image corruptions.

Submission Type: Regular submission (no more than 12 pages of main content)

Assigned Action Editor: ~Animesh_Garg1

Submission Number: 6729