I-sheng Fang
Research Assistant at Academia Sinica, Taiwan
New Taipei, Taiwan
Summary
👤
Senior🎓
Top SchoolI-sheng Fang is a research-focused software engineer specializing in computer vision and generative AI, currently a Research Assistant at Academia Sinica’s CITI working with Dr. Jun-Cheng Chen. He holds an MS in Robotics from National Yang Ming Chiao Tung University and has nine years of research and industry experience spanning academic labs, Microsoft’s Face Science team, and applied roles in PCB segmentation and style transfer. His work blends deep learning, depth estimation, and creative applications like East Asian ideograph font design and typographic style transfer, with a WACV co-first author paper on steganographic stylization. Practical experimentation with ViT vs. ConvNet perceptual backbones at Microsoft highlights his focus on when architecture choices matter for texture- versus context-driven image translation. Outside research he pursues typography and film photography, reflecting a rare mix of technical rigor and visual craft.
9 years of coding experience
2 years of employment as a software developer
Graduate student, Computer Science, 4.3/4.3, Graduate student, Computer Science, 4.3/4.3 at 國立政治大學Bachelor of Science (BS), Applied Mathematics, 2.63 / 4.3, Bachelor of Science (BS), Applied Mathematics, 2.63 / 4.3 at National Chengchi UniversityMaster of Science - MS, Robotics, 4.26/4.3, Master of Science - MS, Robotics, 4.26/4.3 at National Yang Ming Chiao Tung University
Github Skills (18)
perceptual10
machine-learning10
spectral10
datasets10
normalization10
keras10
deep-learning10
tensorflow10
torchvision10
perceptual-losses10
convolution10
computer-vision10
pytorch9
pip8
python7
Programming languages (2)
Github contributions (5)
IShengFang/Self-Contained_Stylization
Aug 2020 - Dec 2022
Contributions:20 commits, 31 pushes, 1 branch in 2 years 3 months
IShengFang/ishengfang
Jul 2020 - Dec 2024
Contributions:65 pushes, 1 branch in 4 years 5 months
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.