1 paper
Yanqi Ge, Jiaqi Liu, Qingnan Fan +6
In this work, we target the task of text-driven style transfer in the context of text-to-image (T2I) diffusion models. The main challenge is consistent structure preservation while…