• Report about this project: Text-Image Pair Generation via Pre-trained Vision-Language Models (Read More) published in EECS595 Fall22 Class Projects Gallery
• In this project, we want to explore about the text2image for the recent popular art AI generator Midjourney AI, and utilize sota vision-language models such as mPlUG and OFA to train on the dataset we created using Midjourney AI, and analyse and compare the performance of these two models.