With the rapid development of face manipulation technology, it is difficult for human eyes to distinguish fake face images. On the contrary, Convolutional Neural Network (CNN) discriminators can quickly reach high accuracy in identifying fake/real face images. In this study, we explore the behavior of CNN models in distinguish fake/real faces. We find multi-scale texture difference information plays an important role in face forgery detection. Motivated by the above observation, we propose a new Multi-scale Texture Difference model coined as MTD-Net for robust face forgery detection, which leverages central difference convolution (CDC) and atrous spatial pyramid pooling (ASPP). CDC combines the pixel intensity information and the pixel gradient information to give a stationary description of texture difference information. Simultaneously, based on the ASPP, multi-scale information fusion can keep the texture features from being destroyed. Experimental results on several databases, Faceforensics++, DeeperForensics-1.0, Celeb-DF and DFDC prove that our MTD-Net outperforms existing approaches. The MTD-Net is more robust to image distortion, e.g., JPEG compression and blur, which is urgently needed in the wild world.

中文翻译：

---

MTD-Net：学习通过多尺度纹理差异检测 Deepfakes 图像


随着人脸操纵技术的快速发展，人眼很难辨别假人脸图像。相反，卷积神经网络（CNN）鉴别器可以快速达到识别假/真人脸图像的高精度。在这项研究中，我们探讨了 CNN 模型在区分假脸/真脸方面的行为。我们发现多尺度纹理差异信息在人脸伪造检测中起着重要作用。受上述观察的启发，我们提出了一种新的多尺度纹理差异模型，称为 MTD-Net，用于鲁棒的人脸伪造检测，该模型利用中心差分卷积（CDC）和多孔空间金字塔池（ASPP）。 CDC结合像素强度信息和像素梯度信息，给出纹理差异信息的平稳描述。同时，基于ASPP的多尺度信息融合可以保证纹理特征不被破坏。在 Faceforensics++、DeeperForensics-1.0、Celeb-DF 和 DFDC 等多个数据库上的实验结果证明，我们的 MTD-Net 优于现有方法。 MTD-Net 对图像失真（例如 JPEG 压缩和模糊）具有更强的鲁棒性，这在现实世界中是迫切需要的。