Siyuan Bian

Student at Shanghai Jiao Tong University MVIG-SJTU

Shanghai, China
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
Siyuan Bian is a master's student at Shanghai Jiao Tong University with five years of hands-on engineering experience focused on machine learning and 3D human pose estimation. He contributed as an ML engineer to the official HybrIK repository (CVPR 2021), improving rotation/quaternion handling, data preprocessing, and integrating camera parameters to boost model reliability. Comfortable debugging complex geometry and transformation pipelines, Siyuan combines research-aware rigor with practical code fixes that materially raise model performance. Based in Shanghai, he brings a pragmatic blend of academic study and open-source impact, favoring precise numerical fixes over sweeping rewrites. Notably, his contributions reveal attention to subtle implementation details—like flip-twist corrections—that often determine real-world success in pose estimation systems.
code5 years of coding experience
github-logo-circle

Github Skills (4)

computer-vision10
pytorch10
pose-estimation9
python9

Programming languages (2)

DartPython

Github contributions (5)

github-logo-circle
jeffffffli/HybrIK

Nov 2021 - Aug 2022

Official code of "HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation", CVPR 2021
Role in this project:
userML Engineer
Contributions:7 commits, 7 PRs, 4 pushes in 9 months
Contributions summary:Siyuan focused on debugging and improving the HybrIK model's performance. Their contributions include fixing issues related to the rotation matrix calculation and quaternion conversions. They also made modifications to the data preprocessing and transformation pipelines by adding new functionalities and fixing the flip twist. The user implemented changes to integrate camera parameters into the training process.
inverse-kinematicscvpr213d-human-pose3d-pose-estimationhuman-pose
jeffffffli/NIKI

May 2023 - Jul 2024

Code of "NIKI: Neural Inverse Kinematics with Invertible Neural Networks for 3D Human Pose and Shape Estimation", CVPR 2023
Contributions:6 PRs, 1 push, 11 comments in 1 year 1 month
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.
Request Free Trial