SparseViT

image-20250610153237395

Su, L., Ma, X., Zhu, X., Niu, C., Lei, Z., and Zhou, J.-Z. 2025. Can We Get Rid of Handcrafted Feature Extractors? SparseViT: Nonsemantics-Centered, Parameter-Efficient Image Manipulation Localization Through Spare-Coding Transformer. AAAI Conference on Artificial Intelligence, 7024–7032.

作者提到现有的

实验

训练集：CAT-Net (Kwon et al. 2021)

测试集：

• CASIAv1 (Dong, Wang, and Tan 2013),

• NIST16 (Guan et al. 2019),

• COVERAGE (Wen et al. 2016),

• Columbia (Hsu and Chang 2006), and

• DEF-12k (Mahfoudi et al. 2019)

像素级别的表现：

image-20250610160655543

image-20250610160752492