Kimberly (Wei-Lin) Hsiao
I work on image generation and editing @ Playground AI
PhD in Computer Science from UT Austin
PhD in Computer Science from UT Austin
I'm currently a Research Scientist at Playground AI, where I work on image generation and image editing.
Before that, I was at Facebook AI Research (FAIR) and Amazon working on various topics in computer vision.
I received my PhD in Computer Science from UT Austin.
University of Texas at Austin
Advisor: Kristen Grauman
2015 - 2021
National Taiwan University
2011 - 2015
Worked with Kristen Grauman on projected related to fashion images (CVPR 2020 & ICCV 2021).
Publications
Worked with Kristen Grauman, Devi Parikh, and Chao-Yuan Wu on Fashion++ (ICCV 2019).
Project details
"Artificial Intelligence System Gives Fashion Advice"
2019 Read Article
Developing state-of-the-art image generation models using multimodal large language models (MLLM) and diffusion models. My work focuses on a unified model for comprehensive image generation and editing.
Developed diffusion models for accurately synthesizing clothing on human images with diverse body shapes.
Developed techniques for improving an outfit's fashion style and creating
synthetic content that makes minmal adjustment to the existing style while improving its fashionability through advanced computer vision algorithms.
Results published in ICCV 2019.
International Conference on Computer Vision (ICCV), 2021
Computer Vision and Pattern Recognition (CVPR), 2018