Higher-Order Ratio Cycles for Fast and Globally Optimal Shape Matching
By Paul Roetzer, Viktoria Ehm, Daniel Cremers, Zorah Lähner, Florian Bernard
Published in IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
Geometry in Machine Learning
Research Group at the University of Bonn and Lamarr Institute
The Geometry in Machine Learning group works at the intersection of machine learning, computer vision and computer graphics. Our research focuses on developing novel techniques that leverage geometric insights to enhance the capabilities of machine learning models. By studying the underlying geometric structures of data, we aim to create more robust, efficient, and interpretable algorithms. Our work spans a wide range of topics, including geometric representation learning, classic geometry processing, and learning under symmetry, as well as applications in physics.
