Jan E. Gerken
Papers
1
Total Citations
8
H-Index
1
About
Dr. Jan E. Gerken is a leading researcher at the intersection of computer vision, robotics, and geometric deep learning, with a particular focus on processing wide-angle and spherical imagery. His most notable contribution is the development of **HEAL-SWIN**, a Vision Transformer architecture designed to operate directly on the sphere, which was published in 2024 and has already garnered 8 citations. This work addresses a critical bottleneck in autonomous driving and robotics: the severe distortion and information loss that occurs when standard convolutional neural networks or vision transformers process fisheye images after projecting them onto a flat plane. By operating natively on spherical data, HEAL-SWIN preserves the geometric integrity of the input, enabling more accurate and robust perception for navigation and scene understanding. Dr. Gerken’s research is pioneering the shift away from planar assumptions in deep learning, offering a principled framework for handling the omnidirectional sensors that are becoming standard in modern robotic systems. His work is essential reading for anyone interested in spherical CNNs, vision transformers for non-Euclidean data, or robust perception for autonomous agents.
Research Focus
Key Achievements
Top Papers
- 1HEAL-SWIN: A Vision Transformer on the Sphere8 citations · 2024