A hybrid minimizing movement and neural network approach to Willmore flow
Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications · 02.03.2026
DOI: 10.4171/ifb/563 ↗Kurz gefasst
Fachartikel in der Zeitschrift Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications, erschienen am 02.03.2026. Verfasst von Martin Rumpf und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).
Abstract (Originalsprache)
We present a hybrid method combining a minimizing movement scheme with neural operators for the simulation of phase field-based Willmore flow. The minimizing movement component is based on a standard optimization problem on a regular grid, whereas the functional to be minimized involves a neural approximation of mean curvature flow proposed by Bretin et al. [J. Comput. Phys. 470 (2022), article no. 111579]. Numerical experiments confirm stability for large time step sizes, consistency, and significantly reduced computational cost compared to a traditional finite element method. Moreover, applications demonstrate its effectiveness in surface fairing and reconstructing of damaged shapes. Thus, our approach offers a robust and efficient tool for geometry processing.
Zitieren
APA
Rumpf, M., Sassen, J., & Smoch, C. (2026). A hybrid minimizing movement and neural network approach to Willmore flow. Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications. https://doi.org/10.4171/ifb/563
Vancouver
Rumpf M, Sassen J, Smoch C. A hybrid minimizing movement and neural network approach to Willmore flow. Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications. 2026. doi:10.4171/ifb/563
BibTeX
@article{Rumpf2026,
author = {Rumpf, Martin and Sassen, Josua and Smoch, Christoph},
title = {A hybrid minimizing movement and neural network approach to Willmore flow},
journal = {Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications},
year = {2026},
doi = {10.4171/ifb/563}
}
EUVIMED