Detecting Generated Images by Real Images Only
This addresses the challenge of authenticating images in real-world scenarios, offering a more efficient and generalizable solution compared to existing methods that lag behind generative models.
The paper tackles the problem of detecting AI-generated images by proposing a method that learns only from real images, mapping them to a dense subspace to distinguish generated images, achieving competitive performance with state-of-the-art methods while using 99.9% less training data and showing robustness against post-processing.
As deep learning technology continues to evolve, the images yielded by generative models are becoming more and more realistic, triggering people to question the authenticity of images. Existing generated image detection methods detect visual artifacts in generated images or learn discriminative features from both real and generated images by massive training. This learning paradigm will result in efficiency and generalization issues, making detection methods always lag behind generation methods. This paper approaches the generated image detection problem from a new perspective: Start from real images. By finding the commonality of real images and mapping them to a dense subspace in feature space, the goal is that generated images, regardless of their generative model, are then projected outside the subspace. As a result, images from different generative models can be detected, solving some long-existing problems in the field. Experimental results show that although our method was trained only by real images and uses 99.9\% less training data than other deep learning-based methods, it can compete with state-of-the-art methods and shows excellent performance in detecting emerging generative models with high inference efficiency. Moreover, the proposed method shows robustness against various post-processing. These advantages allow the method to be used in real-world scenarios.