THB-Diff: a GPU-accelerated differentiable programming framework for THB-splines
Engineering with Computers · 2024-12 · 40 (6) · 3477-3493
DOI: 10.1007/s00366-023-01929-1 ↗Kurz gefasst
Fachartikel in der Zeitschrift Engineering with Computers, online erschienen am 28.12.2023. Verfasst von Ajith Moola und 3 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.
Abstract (Originalsprache)
We have developed a differentiable programming framework for truncated hierarchical B-splines (THB-splines), which can be used for several applications in geometry modeling, such as surface fitting and deformable image registration, and can be easily integrated with geometric deep learning frameworks. Differentiable programming is a novel paradigm that enables an algorithm to be differentiated via automatic differentiation, i.e., using automatic differentiation to compute the derivatives of its outputs with respect to its inputs or parameters. Differentiable programming has been used extensively in machine learning for obtaining gradients required in optimization algorithms such as stochastic gradient descent (SGD). While incorporating differentiable programming with traditional functions is straightforward, it is challenging when the functions are complex, such as splines. In this work, we extend the differentiable programming paradigm to THB-splines. THB-splines offer an efficient approach for complex surface fitting by utilizing a hierarchical tensor structure of B-splines, enabling local adaptive refinement. However, this approach brings challenges, such as a larger computational overhead and the non-trivial implementation of automatic differentiation and parallel evaluation algorithms. We use custom kernel functions for GPU acceleration in forward and backward evaluation that are necessary for differentiable programming of THB-splines. Our approach not only improves computational efficiency but also significantly enhances the speed of surface evaluation compared to previous methods. Our differentiable THB-splines framework facilitates faster and more accurate surface modeling with local refinement, with several applications in CAD and isogeometric analysis.
Förderung
- National Science Foundation
Zitieren
APA
Moola, A., Balu, A., Krishnamurthy, A., & Pawar, A. (2024). THB-Diff: a GPU-accelerated differentiable programming framework for THB-splines. Engineering with Computers, 40(6), 3477-3493. https://doi.org/10.1007/s00366-023-01929-1
Vancouver
Moola A, Balu A, Krishnamurthy A, Pawar A. THB-Diff: a GPU-accelerated differentiable programming framework for THB-splines. Engineering with Computers. 2024;40(6):3477-3493. doi:10.1007/s00366-023-01929-1
BibTeX
@article{Moola2024,
author = {Moola, Ajith and Balu, Aditya and Krishnamurthy, Adarsh and Pawar, Aishwarya},
title = {THB-Diff: a GPU-accelerated differentiable programming framework for THB-splines},
journal = {Engineering with Computers},
year = {2024},
volume = {40},
number = {6},
pages = {3477--3493},
doi = {10.1007/s00366-023-01929-1}
}
EUVIMED