This project is related to the paper "MaSkel: A Model for Human Whole-body X-rays Generation from Human Masking Images". In our work, we did different experiments, here we just listed the MaSkel model we designed and the related training pipeline. As for the augmented method including unconditional image generation and diffusion-based image super-resolution process, we recommend you utilize the open Python package MONAI and the work ResShift.

We also provided two synthetic human x-ray image datasets, because our research was conducted using the UK Biobank Resource, we submitted the data to it and they will release the data after the company's review is completed.