Christoph Heindl
Research Scientist Machine Learning Computer Vision at PROFACTOR GmbH
Steyr, Upper Austria
Summary
🤩
Rockstar🎓
Top SchoolChristoph Heindl is a research scientist specializing in machine learning and computer vision, with 15 years of experience at the intersection of perception, robotics and deep learning. Holding a Dr. techn. (with distinction) from JKU, he focuses on probabilistic methods to generate optimized synthetic training data and builds practical systems that bridge research and deployment. As lead developer and product manager of the widely used real-time 3D scanner ReconstructMe, he has a track record of shipping robust open-source tools adopted by research groups at companies like Facebook, Google and Microsoft. At PROFACTOR he combines hands-on engineering with research, contributing backend and evaluation tooling such as py-motmetrics for benchmarking multi-object tracking. Based in Upper Austria, he pairs rigorous academic training with pragmatic software craftsmanship and an uncommon emphasis on measurement-driven evaluation of perception systems.
15 years of coding experience
Dr. techn. (with distinction), Computer Vision / Machine Learning, Dr. techn. (with distinction), Computer Vision / Machine Learning at Johannes Kepler Universität LinzSoftware Engineering, Software Engineering at University of Applied Sciences Upper Austria - Hagenberg CampusApplied Economics, Applied Economics at Fernuniversität Hagen
German, English, Italian
Github Skills (8)
pandas10
python10
metric10
data-analysis10
numpy9
scipy8
data-science8
pytest7
Programming languages (11)
Github contributions (5)
cheind/py-motmetrics
Apr 2017 - Dec 2022
:bar_chart: Benchmark multiple object trackers (MOT) in Python
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:1 release, 224 commits, 23 PRs in 5 years 9 months
Contributions summary:Christoph contributed to the development and evaluation of metrics for multiple object tracking (MOT) in Python. Their contributions included implementing the core CLEAR MOT metrics, adding support for loading data from MOT challenge datasets, and refactoring distance computations. Furthermore, the user improved the functionality by adding support for fragmentation metrics and ID measures, which are crucial for evaluating tracker performance.
pythontrackerchartbar-chartobject-detection
cheind/py-globalflow
Jul 2021 - Aug 2021
Contributions:59 commits, 65 pushes, 5 branches in 26 days
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.