With the advancement of various GAN and Diffusion technologies, highly realistic synthetic images are becoming increasingly prevalent. The videos or pictures can easily be abused by malicious users to cause severe societal problems or political threats. To mitigate such risks, it’s imperative to find effective ways to detect those fake faces. In this paper, we provide a way of detecting forgery artifacts caused by up-sampling operator in GAN or Diffusion Model. We introduce a method of detecting fake face based on the features from up-sampling, which analyzes neighboring pixels to capture and characterize the generalized structural artifacts from up-sampling technologies. An open-world dataset has been used for the proposed model training and testing. The experimental results indicate that the proposed approach is efficient and effective in detecting up-sampling artifacts.