Paul Roetzer



I am a PhD Student at University of Bonn under the supervision of Prof. Dr. Florian Bernard.

You can contact me via email (e.g. if you have interest in a Bachelor’s or a Master’s thesis).

Please visit my personal webpage for more information.

Bio:

  • B.Sc. Mechanical Engineering at TUM (2015 – 2019)
  • M.Sc. Robotics, Cognition, Intelligence at TUM (2019-2022)

Interests:

  • Optimization
  • Shape Analysis
  • Computer Vision

Publications:

2024

SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency
P. Roetzer; F. Bernard
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
[pdf] selected as oral (3.3% of accepted papers / 0.78% of submissions)
Partial-to-Partial Shape Matching with Geometric Consistency
V. Ehm; M. Gao; P. Roetzer; M. Eisenberger; D. Cremers; F. Bernard
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
[pdf]
Geometrically Consistent Partial Shape Matching
V. Ehm; P. Roetzer; M. Eisenberger; M. Gao; F. Bernard; D. Cremers
In: International Conference on 3D Vision (3DV), 2024.
[pdf]
Revisiting Map Relations for Unsupervised Non-Rigid Shape Matching
D. Cao; P. Roetzer; F. Bernard
In: International Conference on 3D Vision (3DV), 2024.
[pdf]

2023

Structured Prediction Problem Archive
P. Swoboda; A. Abbas; F. Bernard; A. Hornáková; P. Roetzer; B. Savchynskyy
In: arXiv, 2023.
[pdf]
SIGMA: Scale-Invariant Global Sparse Shape Matching
M. Gao; P. Roetzer; M. Eisenberger; Z. Laehner; M. Moeller; D. Cremers; F. Bernard
In: International Conference on Computer Vision (ICCV), 2023.
[pdf]
Unsupervised Learning of Robust Spectral Shape Matching
D. Cao; P. Roetzer; F. Bernard
In: SIGGRAPH / ACM Transactions on Graphics (TOG), 2023.
[pdf] [code] [project page]
Conjugate Product Graphs for Globally Optimal 2D-3D Shape Matching
P. Roetzer; Z. Laehner; F. Bernard
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
[pdf] [code]

2022

P. Roetzer; P. Swoboda; D. Cremers; F. Bernard
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
[pdf] [code]