Scholar
Che-Jui Chang
Google Scholar ID: TepkA_wAAAAJ
Amazon
Citations & Impact
Citations
135
H-index
7
i10-index
6
Publications
16
Co-authors
24
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published multiple papers on topics such as text-to-motion generation, transforming one-line prompts into immersive multi-modal digital stories, and multi-modal dynamic emulation on historical battles; received the Reproducibility Award and Best Appropriateness at ICMI 2022; holds two patents: Body tracking from monocular video (ROBLOX) and Method for converting voice into virtual face image (Red Pill Lab). Also served as a referee for several top international conferences.
Research Experience
- Developed patented ML solutions for real-time lip-sync and motion capture systems at Red Pill Lab; did a research internship at ROBLOX's CoreAI team for real-time body tracking and worked as an Applied Scientist Intern at Amazon on self-supervised representation learning for Selling Partner Services during his PhD.
Education
- Earned a Ph.D. in Computer Science from Rutgers University under Prof. Mubbasir Kapadia; completed an M.S. in EECS and a B.S. in Physics, both from National Taiwan University.
Background
- Works as an Applied Scientist at Amazon Fulfillment Technology (AFT) AI team. Research interests include Computer Vision and Vision-Language Models, focusing on streamlining automation.
Miscellany
- Has given talks on various occasions, covering topics like human motion generation and learning from synthetic data.
