Detecting Deepfakes with Self-Blended Images
Detecting Deepfakes with Self-Blended Images
复制标题
DOI：
10.1109/cvpr52688.2022.01816
复制
发表时间：
2022-04
期刊：
影响因子：
--
通讯作者：
Kaede Shiohara;T. Yamasaki
中科院分区：
文献类型：
--
作者：
Kaede Shiohara;T. Yamasaki
在本文中，我们提出了一种新的被称为自混合图像(SBI)的合成训练数据来检测深伪。SBI是通过混合来自单个原始图像的伪源图像和目标图像，再现常见的伪造伪影(例如，混合边界以及源图像和目标图像之间的统计不一致)来生成的。SBIS背后的关键思想是，更普遍且难以识别的虚假样本鼓励分类器学习通用和健壮的表示，而不过度匹配特定于操作的人工产物。我们通过遵循标准的交叉数据集和交叉操作协议，在FF++、CDF、DFD、DFDC、DFDCP和FFIW数据集上将我们的方法与最先进的方法进行了比较。大量实验表明，该方法提高了模型对未知操作和场景的泛化能力。特别是，在现有方法存在训练集和测试集之间的领域差距的DFDC和DFDCP上，我们的方法在跨数据集评估中分别比基线高4.90%和11.78%。代码可在https://github.com/mapooon/SelfBlendedImages.上找到
In this paper, we present novel synthetic training data called self-blended images (SBIs) to detect deepfakes. SBIs are generated by blending pseudo source and target images from single pristine images, reproducing common forgery artifacts (e.g., blending boundaries and statistical inconsistencies between source and target images). The key idea behind SBIs is that more general and hardly recognizable fake samples encourage classifiers to learn generic and robust representations without overfitting to manipulation-specific artifacts. We compare our approach with state-of-the-art methods on FF++, CDF, DFD, DFDC, DFDCP, and FFIW datasets by following the standard cross-dataset and cross-manipulation protocols. Extensive experiments show that our method improves the model generalization to unknown manipulations and scenes. In particular, on DFDC and DFDCP where existing methods suffer from the domain gap between the training and test sets, our approach outperforms the baseline by 4.90% and 11.78% points in the cross-dataset evaluation, respectively. Code is available at https://github.com/mapooon/SelfBlendedImages.