Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection

Detecting AI-generated images (AIGIs), such as natural images or face images, has become increasingly important yet challenging. In this paper, we start from a new perspective to excavate the reason behind the failure generalization in AIGI detection, named the asymmetry phenomenon, where a naively trained detector tends to favor overfitting to the limited and monotonous fake patterns, causing the feature space to become highly constrained and low-ranked, which is proved seriously limiting the expressivity and generalization. One potential remedy is incorporating the pre-trained knowledge within the vision foundation models (higher-ranked) to expand the feature space, alleviating the model's overfitting to fake. To this end, we employ Singular Value Decomposition (SVD) to decompose the original feature space into two orthogonal subspaces. By freezing the principal components and adapting only the remained components, we preserve the pre-trained knowledge while learning fake patterns. Compared to existing full-parameters and LoRA-based tuning methods, we explicitly ensure orthogonality, enabling the higher rank of the whole feature space, effectively minimizing overfitting and enhancing generalization. We finally identify a crucial insight: our method implicitly learns a vital prior that fakes are actually derived from the real, indicating a hierarchical relationship rather than independence. Modeling this prior, we believe, is essential for achieving superior generalization. Our codes are publicly available at https://github.com/YZY-stack/Effort-AIGI-Detection.

In recent years, the rise of AI-generated images (AIGIs) has created both excitement and challenges. One major issue we noticed is that many detection methods struggle to identify the fake images effectively in the field of security (e.g. face recognition system). This happens because they often focus too much on a limited set of fake patterns neglecting the importance of real information, which makes it hard for them to recognize new or different types of fakes. To address this problem, we start from the perspective of how to distinguish fakes while learning good real information. Specifically, we break down the detection model into two parts: the important ones keeping real images's information from existing advanced models and the others that can adapt to identify fakes. With this fresh approach, we improved the model's ability to recognize a wider variety of AI-generated images.Our research not only improves the detection capacity but also highlights that the fakes often come from real images, suggesting a significant connection between them. Understanding this relationship is key to creating more stable detection systems to ensure the responsible use of AI-generated content.