ForgeryNet 人脸伪造数据集
收藏资源简介:
ForgeryNet 数据集是一个庞大且全面的基准测试,专为深度伪造分析而构建。它包含了 290 万张图像和 221,247 个视频,涵盖了来自全球的 7 种图像层面和 8 种视频层面的伪造操作方法。这个数据集为研究者提供了丰富的资源,以支持图像和视频层面的四种任务:图像伪造分类、空间伪造定位、视频伪造分类和时间伪造定位。这些任务包括了从二分类到多分类的图像伪造识别,以及对伪造区域的空间和时间定位。
ForgeryNet is a large-scale, comprehensive benchmark specifically built for deepfake analysis. It contains 2.9 million images and 221,247 videos, covering 7 image-level and 8 video-level forgery manipulation techniques sourced from across the globe. This dataset offers researchers abundant resources to support four tasks spanning both image and video domains: image forgery classification, spatial forgery localization, video forgery classification, and temporal forgery localization. These tasks encompass binary-to-multi-class image forgery recognition, as well as spatial and temporal localization of forged regions.




