Jianzhu Guo
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolJianzhu Guo is a research-focused software engineer with 11 years of experience specializing in 3D face analysis, face recognition, and GenAI-driven human modeling. He earned a Ph.D. from the National Laboratory of Pattern Recognition (CAS) under Stan Z. Li and Zhen Lei, and now builds production-facing GenAI features for human-centric applications at Kwai Technology. His open-source work—most notably the widely used 3DDFA and 3DDFA_V2 PyTorch implementations—demonstrates deep expertise in 3D dense face alignment, rasterization, and practical model refactoring. He blends research rigor with engineering pragmatism, contributing core algorithm improvements, visualization tools, and benchmarks that bridge academia and product. Based in Haidian, Beijing, he publishes code and models on GitHub and HuggingFace and is active in making 3D face technology accessible and fun through projects like LivePortrait.
11 years of coding experience
4 years of employment as a software developer
博士, 模式识别与人工智能, 博士, 模式识别与人工智能 at 中国科学院自动化研究所Geographic Information System, Geographic Information System at 东南大学
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (14)
computer-vision10
pytorch10
deep-learning10
python10
3d10
rasterizing9
raster9
rasterizer9
rasterize9
ras9
opencv9
c-language8
cprogramming-language8
relu7
Programming languages (10)
Github contributions (5)
cleardusk/3DDFA_V2
Aug 2020 - Oct 2021
The official PyTorch implementation of Towards Fast, Accurate and Stable 3D Dense Face Alignment, ECCV 2020.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:3 releases, 107 commits, 2 PRs in 1 year 1 month
Contributions summary:Jianzhu primarily updated and refined code related to 3D face alignment, model updates and refactoring. The user worked on the core code, including modifying the rasterization functions and adding new features like PNCC, depth, and UV texture spanning. They also added support for different models, including a resnet backbone and various visualization tools, demonstrating a focus on improving the project's functionality and output.
pytorch3dmm3d-face-alignmenteccvface-alignment
cleardusk/3DDFA
Jun 2018 - May 2022
The PyTorch improved version of TPAMI 2017 paper: Face Alignment in Full Pose Range: A 3D Total Solution.
Role in this project:ML Engineer
ML EngineerContributions:1 release, 157 commits, 10 PRs in 3 years 11 months
Contributions summary:Jianzhu primarily contributed to the development and improvement of a 3D face alignment model within the repository. They made several updates to the training process, including modifications to loss functions and learning rate adjustments. Additionally, the user added a benchmark pipeline for evaluating the model's performance and implemented a depth rendering feature.
pythonimprovedtotalface-alignmentcomputer-vision
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.