Sven Meier
Manager Robotics at Toyota Motor Europe
Beijing, Belgium
Summary
👤
Senior🎓
Top SchoolSven Meier is a robotics manager based in Beijing with 10 years of experience applying machine learning, statistical modelling and digital image/signal processing to real-world problems. He leads Toyota Motor Europe’s robotics team, turning remote sensing and algorithm research into deployable computer vision solutions for societal and business challenges. Sven holds a PhD in radiometric remote sensing and an MEng in electronic and electrical engineering, blending deep academic expertise with product-focused engineering. An active open-source contributor, he has improved rendering for 6D object pose benchmarks and fixed robustness issues in Detectron2, showing attention to both research-grade tooling and production frameworks. Colleagues value his knack for bridging complex signal-processing theory with pragmatic systems engineering—and he humorously flags his persistence with “permanent head damage” on GitHub.
10 years of coding experience
Doctor of Philosophy (Ph.D.), Radiometric remote sensing, Doctor of Philosophy (Ph.D.), Radiometric remote sensing at University of StrathclydeEngineer's Degree, Electronic and Electrical Engineering, MEng (hons.), Engineer's Degree, Electronic and Electrical Engineering, MEng (hons.) at University of Bath
English, German, French
Github Skills (12)
computer-vision10
pytorch10
3d-rendering10
opengl10
python10
object-detection9
scripting9
image-segmentation8
segmentation8
machine-learning7
deep-learning7
data-science6
Programming languages (15)
Github contributions (5)
thodan/bop_toolkit
Nov 2021 - Nov 2021
A Python toolkit of the BOP benchmark for 6D object pose estimation.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:4 reviews, 8 commits, 3 PRs in 14 days
Contributions summary:Sven implemented a Vispy-based renderer for the BOP toolkit, allowing for offscreen rendering with Python. This involved adding a new file, `renderer_vispy.py`, that utilizes Vispy and OpenGL to render 3D object models. Furthermore, the user updated several scripts and the main renderer file to incorporate and utilize the new Vispy renderer, ensuring its integration within the existing framework. This update demonstrates the user's contribution to improving rendering capabilities for the BOP benchmark.
object-pose-estimationpythonpose-estimationbenchmarkestimation
facebookresearch/detectron2
Oct 2019 - Jun 2020
Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.
Role in this project:Backend Developer
Backend DeveloperContributions:5 commits, 9 PRs, 19 comments in 8 months
Contributions summary:Sven primarily contributed to bug fixes and minor enhancements within the Detectron2 framework. Their work involved correcting a typo in dataset documentation, resolving a type error issue in the visualizer module, updating logging messages related to dataset statistics, and making adjustments to allow for tensor inputs for mask pasting. Additionally, the user addressed an issue by allowing the `convert_to_coco_dict` function to work with custom datasets that lacked annotations.
detectron2deep-learningface-recognitionrecognitionobject-detection
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.