官方服务：
资源简介：
dataset for image forgery
应用场景：
创建时间：
2025-06-05
相关数据集
NeXT-IMDL
NeXT-IMDL是由清华大学团队构建的大规模诊断性基准数据集，旨在系统评估下一代图像篡改检测与定位模型的泛化能力。该数据集包含55.8万样本，覆盖32种主流生成模型（如Stable Diffusion、FLUX及商业工具PS等）创建的篡改内容，通过掩码、文本和参考图像等多模态引导条件生成。数据来源包括COCO等公开图像库，采用四维度分类框架（编辑模型、篡改类型、语义标签、篡改粒度）构建，重点解决
arXiv2025-12-29 更新580
PS-Battles
PS-Battles The PS-Battles dataset, an image collection for image manipulation detection. Download instructions Simply put the download.py script and the originals.tsv
Zenodo2021-03-12 更新80
Comparison experiments.
The forensic examination of AIGC(Artificial Intelligence Generated Content) faces poses a contemporary challenge within the realm of color image forensics. A myriad of artificially generated faces by
Figshare2025-01-28 更新30
StegoAppDB Database Sample
This dataset is a subset of the full StegoAppDB database (Newman 2019) and contains 51,200 images from 28 mobile phones across 10 models. Student photographers captured 100 high-resolution images per
DataCite Commons2026-02-11 更新120
CG-1050: Original and tampered images (Color and grayscale)
This dataset is composed of 100 natural images, 1050 tampered counterparts, and 1380 masks. For every image were applied from 10 to 11 manipulations among copy-move, cut-paste, colorizing and retouch
Mendeley Data60
