In the computer vision domain, diffusion models have significantly advanced the image synthesis quality in generative tasks. This talk will introduce two pieces of my recent research on diffusion-based image generation and translation. In detail, I will present (1) MagiCapture, a personalized portrait image synthesis approach, and (2) StableVITON, a virtual try-on image synthesis given a person's body image and a clothing image. Afterwards, I will discuss conclusions and future research directions.

Jaegul Choo is currently an associate professor in the Kim Jaechul Graduate School of Artificial Intelligence at KAIST. His primary research interest lies in deep learning applications in computer vision and natural language processing. He has been an assistant professor in the Dept. of Computer Science and Engineering at Korea University from 2015 to 2019 and then an associate professor in the Dept. of Artificial Intelligence at Korea University in 2019. He received M.S in the School of Electrical and Computer Engineering at Georgia Tech in 2009 and Ph.D in the School of Computational Science and Engineering at Georgia Tech in 2013.