Philipp Lindenberger
PhD Candidate
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolPhilipp Lindenberger is a PhD candidate in computer science at ETH Zurich with six years of experience bridging computational science, materials engineering and applied machine learning. He contributes to high-impact open-source SfM tooling—improving pycolmap bindings and adding CUDA-enabled feature matching—and helped deliver the ICCV 2021 Best Student Paper project on pixel-perfect structure-from-motion by hardening optimizers, fixing memory leaks and managing GPU/CPU model transfers. His background in materials science and numerical PDEs informs a pragmatic approach to simulation, optimization and low-level performance work, from Fortran UMATs for Abaqus to Python and CUDA pipelines. Philipp has research experience at Google and diverse industry internships, and is notable for combining deep scientific rigor with hands-on engineering across production and research codebases.
6 years of coding experience
2 years of employment as a software developer
University Entrance Degree, Civil Engineering, 1.5, University Entrance Degree, Civil Engineering, 1.5 at HTL Saalfelden (secondary technical school)Bachelor's degree, Material Science, 1.3 (BSc with distinction), Bachelor's degree, Material Science, 1.3 (BSc with distinction) at University of LeobenDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at ETH Zürich
German, English
Github Skills (15)
3d-reconstruction10
structure-from-motion10
computer-vision10
c-language10
deep-learning10
cprogramming-language10
python10
image-processing9
pybind119
3d9
feature-extraction9
cuda9
pytorch9
hdf8
parallel-computing8
Programming languages (3)
Github contributions (5)
cvg/pixel-perfect-sfm
Jan 2022 - Oct 2022
Pixel-Perfect Structure-from-Motion with Featuremetric Refinement (ICCV 2021, Best Student Paper Award)
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 48 commits, 31 PRs in 8 months
Contributions summary:Philipp primarily contributed to the core functionality of the pixel-perfect structure-from-motion project. They focused on fixing and improving the parallel optimizer, addressing memory leaks, and resolving issues related to the handling of different feature types, particularly with dense features. Additionally, they added and updated configurations for various feature extraction methods. Furthermore, the user worked on freeing GPU resources by transferring models between CPU and GPU.
structure-from-motionpytorchiccvpixel-perfectfeature-matching
colmap/pycolmap
Jan 2022 - Oct 2022
Python bindings for COLMAP
Role in this project:Back-end Developer
Back-end DeveloperContributions:68 reviews, 18 commits, 18 PRs in 9 months
Contributions summary:Philipp made significant contributions to the Python bindings for COLMAP. Their work focused on binding various core components such as reconstruction, images, and cameras. They implemented improvements to error handling with exceptions and added functionality for projecting points and aligning reconstructions. Further contributions include CUDA support for feature matching and MVS steps.
structure-from-motionpythonpython-bindingsintrinsicsmotion-tracking
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.