Guardado en:
Detalles Bibliográficos
Autores principales: Bai, Yang, Eskandar, George, Liu, Ziyuan, Kutyniok, Gitta
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2603.15627
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908890751303680
author Bai, Yang
Eskandar, George
Liu, Ziyuan
Kutyniok, Gitta
author_facet Bai, Yang
Eskandar, George
Liu, Ziyuan
Kutyniok, Gitta
contents Traditional fluid dynamics simulation pipelines combine numerical solvers with rendering, producing highly realistic results but at considerable computational cost. Diffusion-based generative video models offer a faster alternative, yet often ignore physical laws and thus fail to capture consistent dynamics. We propose a physics-informed video diffusion framework that jointly generates visual outputs and physical states. Unlike prior two-stage approaches that first simulate the physical variables and then render, we directly integrate physics constraints into the generative process, enabling simultaneous prediction of physical states and realistic videos without a separate rendering step. Built on the two-dimensional shallow water equations with terrain topography, our method produces temporally coherent water flow while maintaining physical plausibility. Experiments show that it outperforms purely data-driven video diffusion baselines in both realism and physical fidelity, while generating videos significantly faster than traditional simulation-plus-rendering pipelines.
format Preprint
id arxiv_https___arxiv_org_abs_2603_15627
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Physics-Informed Video Diffusion For Shallow Water Equations
Bai, Yang
Eskandar, George
Liu, Ziyuan
Kutyniok, Gitta
Graphics
Computational Engineering, Finance, and Science
Computational Physics
Fluid Dynamics
Traditional fluid dynamics simulation pipelines combine numerical solvers with rendering, producing highly realistic results but at considerable computational cost. Diffusion-based generative video models offer a faster alternative, yet often ignore physical laws and thus fail to capture consistent dynamics. We propose a physics-informed video diffusion framework that jointly generates visual outputs and physical states. Unlike prior two-stage approaches that first simulate the physical variables and then render, we directly integrate physics constraints into the generative process, enabling simultaneous prediction of physical states and realistic videos without a separate rendering step. Built on the two-dimensional shallow water equations with terrain topography, our method produces temporally coherent water flow while maintaining physical plausibility. Experiments show that it outperforms purely data-driven video diffusion baselines in both realism and physical fidelity, while generating videos significantly faster than traditional simulation-plus-rendering pipelines.
title Physics-Informed Video Diffusion For Shallow Water Equations
topic Graphics
Computational Engineering, Finance, and Science
Computational Physics
Fluid Dynamics
url https://arxiv.org/abs/2603.15627