FluidFormer: Transformer with Continuous Convolution for Particle-based Fluid Simulation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Nianyi, Chen, Yu, Zheng, Shuai
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912516800512000
author Wang, Nianyi
Chen, Yu
Zheng, Shuai
author_facet Wang, Nianyi
Chen, Yu
Zheng, Shuai
contents Learning-based fluid simulation networks have been proven as viable alternatives to traditional numerical solvers for the Navier-Stokes equations. Existing neural methods follow Smoothed Particle Hydrodynamics (SPH) frameworks, which inherently rely only on local inter-particle interactions. However, we emphasize that global context integration is also essential for learning-based methods to stabilize complex fluid simulations. We propose the first Fluid Attention Block (FAB) with a local-global hierarchy, where continuous convolutions extract local features while self-attention captures global dependencies. This fusion suppresses the error accumulation and models long-range physical phenomena. Furthermore, we pioneer the first Transformer architecture specifically designed for continuous fluid simulation, seamlessly integrated within a dual-pipeline architecture. Our method establishes a new paradigm for neural fluid simulation by unifying convolution-based local features with attention-based global context modeling. FluidFormer demonstrates state-of-the-art performance, with stronger stability in complex fluid scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2508_01537
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FluidFormer: Transformer with Continuous Convolution for Particle-based Fluid Simulation
Wang, Nianyi
Chen, Yu
Zheng, Shuai
Computational Engineering, Finance, and Science
Graphics
Machine Learning
Fluid Dynamics
Learning-based fluid simulation networks have been proven as viable alternatives to traditional numerical solvers for the Navier-Stokes equations. Existing neural methods follow Smoothed Particle Hydrodynamics (SPH) frameworks, which inherently rely only on local inter-particle interactions. However, we emphasize that global context integration is also essential for learning-based methods to stabilize complex fluid simulations. We propose the first Fluid Attention Block (FAB) with a local-global hierarchy, where continuous convolutions extract local features while self-attention captures global dependencies. This fusion suppresses the error accumulation and models long-range physical phenomena. Furthermore, we pioneer the first Transformer architecture specifically designed for continuous fluid simulation, seamlessly integrated within a dual-pipeline architecture. Our method establishes a new paradigm for neural fluid simulation by unifying convolution-based local features with attention-based global context modeling. FluidFormer demonstrates state-of-the-art performance, with stronger stability in complex fluid scenarios.
title FluidFormer: Transformer with Continuous Convolution for Particle-based Fluid Simulation
topic Computational Engineering, Finance, and Science
Graphics
Machine Learning
Fluid Dynamics
url https://arxiv.org/abs/2508.01537