In this paper, some experimental studies on detecting the post-processed GAN-generated face images find that: (a) both the luminance component and chrominance components play an important role; (b) the RGB and YCbCr color spaces achieve better performance than the HSV and Lab color spaces. Therefore, in order to enhance the robustness, both the luminance component and chrominance components of dual-color spaces (RGB and YCbCr) are considered to utilize color information effectively. Besides, convolutional block attention module and multi-layer feature aggregation module are introduced into the Xception model to enhance its representation power of the feature map and aggregate multi-layer features, respectively. Experimental results demonstrate that our method outperforms some existing methods, especially in robustness against different types of post-processing operations, such as JPEG compression, Gaussian blurring, Gamma correction, and so on.
[Chen B, Xin L, et al. A robust GAN-generated face detection method based on dual-color spaces and an improved Xception. IEEE Transactions on Circuits and Systems for Video Technology, 2021.DOI (identifier) 10.1109/TCSVT.2021.3116679]