Mårten Björkman
Professor at KTH Royal Institute of Technology
Greater Stockholm Metropolitan Area Sweden
Summary
👤
Senior🎓
Top SchoolMårten Björkman is a professor and researcher with 13 years of experience specializing in computer vision, machine learning and robotics, currently based in the Greater Stockholm area. He focuses on robot self-modeling—enabling machines to learn their own embodiment and predict the effects of their actions—combining deep reinforcement learning with Gaussian processes. His hands-on background includes high-performance implementations such as a CUDA SIFT port that accelerates feature extraction on NVidia GPUs, reflecting a blend of theoretical insight and low-level optimization skill. Holding a PhD from KTH, he brings decades-long academic depth together with practical system building that spans from GPU kernels to embodied AI.
12 years of coding experience
PhD, Computer Vision, PhD, Computer Vision at KTH Royal Institute of TechnologyMSc, Computer Science & Engineering, MSc, Computer Science & Engineering at The Faculty of Engineering at Lund University
English, Swedish
Github Skills (12)
cuda10
machine-vision10
nvidia10
computer-vision10
gpu10
sift10
c-language8
algorithms8
data-structures8
algorithm8
data-structure8
cprogramming-language8
Programming languages (2)
Github contributions (5)
Celebrandil/CudaSift
Jun 2013 - May 2019
A CUDA implementation of SIFT for NVidia GPUs (1.2 ms on a GTX 1060)
Role in this project:
Back-end Developer
Back-end DeveloperContributions:71 commits, 2 PRs, 99 pushes in 6 years
Contributions summary:Mårten appears to be responsible for the initial development of the CUDA-based SIFT implementation. Their primary contribution is the creation of the core CUDA code, including kernels for image scaling, subtraction, and descriptor extraction. The user also includes modifications to optimize performance, likely targeting specific hardware architectures. This work directly implements the SIFT algorithm on the GPU for faster feature extraction.
cudanvidiasiftgpugtx
EnTimeMent/EnTimeMent.github.io
Feb 2022 - Feb 2022
Contributions:20 pushes, 1 branch in 8 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.