Cheng-han Lee
PhD Candidate, Graduate Research Assistant
Austin, Texas, United States
Summary
🤩
Rockstar🎓
Top SchoolCheng-han Lee is a PhD candidate in ECE at UT Austin and a camera algorithm engineer with a decade of experience building and optimizing video and image ML systems for mobile, cloud, and research products. He has shipped production-facing camera features and NPU-optimized models at MediaTek and Perfect Corp., contributed to large-scale facial datasets and GAN stabilization at SenseTime, and investigated video reframing, ROI tracking, and synthesis in academic work under Prof. Alan Bovik. Recent internships at Amazon’s Astro team and Qualcomm reflect a practical focus on multimodal video understanding and real-time video enhancement. Cheng-han combines deep research credentials—reviewing top-tier CV/ML conferences—with hands-on deployment skills across pruning, NAS, and compiler/toolkit integration, making him adept at closing the gap between state-of-the-art research and efficient on-device inference. An interesting thread through his career is repeatedly bringing high-resolution generative and tracking methods down to resource-constrained hardware without sacrificing visual quality.
10 years of coding experience
5 years of employment as a software developer
Bachelor's degree, Electronic and Computer Engineering, Bachelor's degree, Electronic and Computer Engineering at National Chiao Tung UniversityMaster's degree, Computer Science, 4.0/4.3, Master's degree, Computer Science, 4.0/4.3 at National Taiwan UniversityDoctor of Philosophy - PhD, Electrical and Computer Engineering, 3.9/4.0, Doctor of Philosophy - PhD, Electrical and Computer Engineering, 3.9/4.0 at 美國德州大學奧斯汀分校
Chinese, English
Github Skills (13)
face-detection10
computer-vision10
face10
face-generation10
parsing10
editing10
generative-adversarial-network10
face-recognition10
large-scale9
scale9
super-resolution7
tensorflow3
neural-network2
Programming languages (1)
Github contributions (5)
switchablenorms/CelebAMask-HQ
Apr 2019 - May 2020
A large-scale face dataset for face parsing, recognition, generation and editing.
Contributions:33 commits, 40 pushes, 56 comments in 1 year 1 month
celebaface-datasetface-recognitionface-segmentationgenerative-adversarial-network
steven413d/AACNN
Dec 2018 - Nov 2019
Tensorflow Inplementation for AACNN : Attribute Augmented Convolutional Neural Network for Face Hallucination (NTIRE2018)
Contributions:16 pushes, 1 branch, 7 comments in 11 months
convolutional-neural-networkconvolutionalsuper-resolutionaugmentedneural-network
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.