Abstract
DeepFaceLab is an open-source deepfake system created by \textbf{iperov} forface swapping with more than 3,000 forks and 13,000 stars in Github: itprovides an imperative and easy-to-use pipeline for people to use with nocomprehensive understanding of deep learning framework or with modelimplementation required, while remains a flexible and loose coupling structurefor people who need to strengthen their own pipeline with other featureswithout writing complicated boilerplate code. In this paper, we detail theprinciples that drive the implementation of DeepFaceLab and introduce thepipeline of it, through which every aspect of the pipeline can be modifiedpainlessly by users to achieve their customization purpose, and it's noteworthythat DeepFaceLab could achieve results with high fidelity and indeedindiscernible by mainstream forgery detection approaches. We demonstrate theadvantage of our system through comparing our approach with current prevailingsystems. For more information, please visit:https://github.com/iperov/DeepFaceLab/.