I'm a computer scientist focusing on computer vision and machine learning. Recently, I've enjoyed thinking about how symmetries in data relate to symmetries in algorithms and I've applied ideas from geometric/equivariant deep learning to computer vision tasks such as image matching.

I completed my PhD at Chalmers in 2024 under the supervision of Fredrik Kahl, and co-advisors Michael Felsberg and Axel Flinth.

Selected publications

Flopping for FLOPs: Leveraging equivariance for computational efficiency
ICML 2025 (Spotlight), arXiv:2502.05169
Georg Bökman, David Nordström, Fredrik Kahl

Steerers: A framework for rotation equivariant keypoint description
CVPR 2024 (Oral), arXiv:2312.02152
Georg Bökman, Johan Edstedt, Michael Felsberg, Fredrik Kahl

Investigating how ReLU-networks encode symmetries
NeurIPS 2023, arXiv:2305.17017
Georg Bökman, Fredrik Kahl