Xinshuo Weng
Senior Research Scientist at NVIDIA
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolXinshuo Weng is a Senior Research Scientist at NVIDIA with a decade of experience bridging 3D computer vision, autonomous driving, and generative models. A CMU Robotics PhD alum and Qualcomm Fellowship recipient, he has advanced end-to-end AV systems, traffic simulation, online mapping, and LLM/VLM integration for vehicles. His work spans research and engineering—contributions include practical fixes and evaluation improvements to the widely used AB3DMOT 3D multi-object tracking codebase. Prior roles at Oculus and Meta focused on facial, body and hand pose estimation, reflecting a strong track record in human-centric perception as well as vehicle-centric autonomy. Based in Toronto, he combines deep academic training with product-facing research that ships in large-scale systems.
10 years of coding experience
3 years of employment as a software developer
B.S. in Electrical Engineering, B.S. in Electrical Engineering at Wuhan UniversityPh.D. in Robotics Computer Vision and Machine Learning, Ph.D. in Robotics Computer Vision and Machine Learning at Carnegie Mellon University School of Computer Science
Chinese, Japanese, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (3)
python10
computer-vision10
machine-learning9
Programming languages (4)
Github contributions (5)
xinshuoweng/AB3DMOT
Jun 2019 - Mar 2022
(IROS 2020, ECCVW 2020) Official Python Implementation for "3D Multi-Object Tracking: A Baseline and New Evaluation Metrics"
Role in this project:
ML Engineer
ML EngineerContributions:216 commits, 30 PRs, 196 pushes in 2 years 9 months
Contributions summary:Xinshuo primarily contributed to the `evaluation/evaluate_kitti3dmot.py` file, indicating a focus on evaluating 3D multi-object tracking performance. Their commits included modifications to the evaluation metrics, addition of results for pedestrian and cyclist classes, and support for Python 3.5 compatibility. Furthermore, they fixed orientation correction bugs within the `main.py` file and updated the example usage within the `trk_conf_threshold.py` file.
pythonkitti3d-mot3d-multi-object-trackingreal-time
xinshuoweng/Mono3DPLiDAR
Jun 2019 - Jun 2020
Contributions:15 commits, 14 pushes, 1 branch in 11 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.