Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection

AI-generated images (AIGIs), such as natural or face images, have 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 $\mathit{\text{asymmetry}}\mathit{\text{phenomenon}}$, 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 $\mathit{\text{two}}\mathit{\text{orthogonal}}\mathit{\text{subspaces}}$. 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 $\mathit{\text{a}}\mathit{\text{vital}}\mathit{\text{prior}}\mathit{\text{that}}\mathit{\text{fakes}}\mathit{\text{are}}\mathit{\text{actually}}\mathit{\text{derived}}\mathit{\text{from}}\mathit{\text{the}}\mathit{\text{real}}$, indicating a hierarchical relationship rather than independence. Modeling this prior, we believe, is essential for achieving superior generalization. Our codes are publicly available at \href{https://github.com/YZY-stack/Effort-AIGI-Detection}{GitHub}.