Zhixuan LIU

I'm a


About Me

I'm Zhixuan Liu and

Hi! I am a fourth-year undergraduate student majoring in Computer Science from School of Data Science, the Chinese University of Hong Kong, Shenzhen. I am also a research assistant and an upcoming master student at Robotics Institute, Carnegie Mellon University, advised by Dr. Jean Oh and PhD student Peter Schaldenbrand. My research interests lie in the intersection of machine learning and creating art, such as painting, video generation, and auto music creation.

Personally, I am a piano player with 15 years of performance experience, and I play with guitar and synthesizer in my spare time as well. I am also a Chinese traditional dancer with two gold awards from Beijing in 2010 and Hong Kong in 2011, and I'm practicing KPOP dance recently. Moreover, I really enjoy taking photos, especially film photography.


Selected Projects

  • Style-CLIPDraw

    description update..

  • SongBot: Auto Music Generator

    description update..

  • Habiter-OneForAll

    description update..

To know more about my projects? Zhixuan's GitHub



  • StyleCLIPDraw: Coupling Content and Style in Text-to-Drawing Synthesis

    The 2021 NeurIPS Workshop on Machine Learning for Creativity and Design

    Accepted for Oral Presentation arXiv Github What's AI on YouTube

    Peter Schaldenbrand, Zhixuan Liu, Jean Oh

    StyleCLIPDraw is a text-to-drawing synthesis model with artistic control via a given style image and content control via a language description.

  • Gander: A Comprehensive Machine Learning Media Synthesis Platform & Style-CLIPDraw: A Style-Transferred Text-to-Drawing Synthesis Method

    CMU RISS Journal 2021

    Paper Video Poster

    Zhixuan Liu, Peter Schaldenbrand, Jean Oh

    Add description

  • SongBot: An Interactive Music Generation Robotic System for Non-musicians Learning from a Song

    IEEE RCAR 2021

    IEEE Xplore Video

    Kaiwen Xue, Zhixuan Liu, Jiaying Li, Xiaoqiang Ji and Huihuan Qian

    Add description


Zhixuan's Life