Matthias Niessner
Co-Founder CEO at SpAItial AI
Greater Munich Metropolitan Area Germany
Summary
🤩
Rockstar🎓
Top SchoolMatthias Niessner is a technology entrepreneur and academic with 12 years of experience building real-time 3D reconstruction and AI-driven media systems from research prototypes to commercial products. As Co-Founder/CEO of SpAItial AI and Co-Founder of Synthesia, he bridges cutting-edge computer vision research with product and company leadership while holding a professorship at Technical University Munich. His PhD-trained expertise in CUDA and C++ is reflected in influential open-source work on VoxelHashing and BundleFusion (SIGGRAPH papers), contributing core depth-sensing and GPU kernels for large-scale, real-time 3D reconstruction. He combines rigorous academic publication pedigree with hands-on systems engineering and startup execution, often working at the intersection of graphics, perception, and scalable production systems. Based in the Greater Munich area, he has a track record of translating complex research—down to optimized GPU kernels—into deployable technologies.
12 years of coding experience
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at FAU Erlangen-Nürnberg
Github Skills (13)
cuda10
kernel10
c-language10
parallel-computing10
cprogramming-language10
image-processing10
3d-reconstruction10
computer-vision9
file-access8
machine-learning8
file-processing8
file-handling8
fileio8
Programming languages (4)
Github contributions (5)
niessner/VoxelHashing
Jun 2014 - Feb 2017
[Siggraph Asia 2013] Large-Scale, Real-Time 3D Reconstruction
Role in this project:
Back-end Developer
Back-end DeveloperContributions:40 commits, 18 pushes in 2 years 8 months
Contributions summary:Matthias primarily contributed to the `DepthSensing` and related CUDA code, suggesting a focus on the back-end processing of 3D reconstruction. They made changes to include and modify core functions for depth sensing and point cloud I/O. The contributions involve the use of C++ and CUDA, as the user worked on features like binary dump loading, rigid transform calculations, and integrating depth maps.
pytorchreal-timereconstructionsiggraph-asia3d-models
niessner/BundleFusion
Sep 2015 - Oct 2016
[Siggraph 2017] BundleFusion: Real-time Globally Consistent 3D Reconstruction using Online Surface Re-integration
Role in this project:Back-end Developer
Back-end DeveloperContributions:87 commits in 1 year 1 month
Contributions summary:Matthias implemented and modified CUDA kernels for image processing tasks within the BundleFusion project. Their work included developing and optimizing kernels for tasks such as image resampling, converting depth maps to camera space positions, and integrating depth data. They also contributed to the CUDAImageManager class, handling image data processing and memory management. The user demonstrated expertise in CUDA programming and optimization techniques, critical to the project's real-time 3D reconstruction capabilities.
pytorchreconstructionstructure-from-motioncomputer-vision3d-reconstruction
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.