Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yan, Zhiyuan, Wang, Jiangming, Jin, Peng, Zhang, Ke-Yue, Liu, Chengchun, Chen, Shen, Yao, Taiping, Ding, Shouhong, Wu, Baoyuan, Yuan, Li
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909617255088128
author Yan, Zhiyuan
Wang, Jiangming
Jin, Peng
Zhang, Ke-Yue
Liu, Chengchun
Chen, Shen
Yao, Taiping
Ding, Shouhong
Wu, Baoyuan
Yuan, Li
author_facet Yan, Zhiyuan
Wang, Jiangming
Jin, Peng
Zhang, Ke-Yue
Liu, Chengchun
Chen, Shen
Yao, Taiping
Ding, Shouhong
Wu, Baoyuan
Yuan, Li
contents 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 \textit{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 \textit{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 \textit{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 \href{https://github.com/YZY-stack/Effort-AIGI-Detection}{GitHub}.
format Preprint
id arxiv_https___arxiv_org_abs_2411_15633
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection
Yan, Zhiyuan
Wang, Jiangming
Jin, Peng
Zhang, Ke-Yue
Liu, Chengchun
Chen, Shen
Yao, Taiping
Ding, Shouhong
Wu, Baoyuan
Yuan, Li
Computer Vision and Pattern Recognition
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 \textit{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 \textit{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 \textit{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 \href{https://github.com/YZY-stack/Effort-AIGI-Detection}{GitHub}.
title Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.15633