Neural Refractive Index Primitives for Flame Field Reconstruction Using Background-Oriented Schlieren

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Xinyi, Hu, Wei, Liao, Zizhou, Wang, Zheng, Zhang, Yue, Li, Jingxuan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911683097657344
author Lu, Xinyi
Hu, Wei
Liao, Zizhou
Wang, Zheng
Zhang, Yue
Li, Jingxuan
author_facet Lu, Xinyi
Hu, Wei
Liao, Zizhou
Wang, Zheng
Zhang, Yue
Li, Jingxuan
contents An improved neural refractive-index-primitive method for background-oriented schlieren tomography is presented, enabling continuous three-dimensional reconstruction of refractive-index fields using a compact multilayer perceptron. The method adopts the refractive-index field as the sole neural primitive and integrates multiresolution hash encoding, automatic-discrete gradient losses, and a three-dimensional mask to enable fast convergence and high-resolution, spatially coherent reconstructions. Tests on numerical combustion phantoms and real flame data demonstrate accurate recovery of both large-scale structures and fine-scale turbulence, strong robustness to noise, and clear advantages over frequency-encoding-based and voxel-based reconstruction methods.
format Preprint
id arxiv_https___arxiv_org_abs_2605_11454
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Neural Refractive Index Primitives for Flame Field Reconstruction Using Background-Oriented Schlieren
Lu, Xinyi
Hu, Wei
Liao, Zizhou
Wang, Zheng
Zhang, Yue
Li, Jingxuan
Fluid Dynamics
An improved neural refractive-index-primitive method for background-oriented schlieren tomography is presented, enabling continuous three-dimensional reconstruction of refractive-index fields using a compact multilayer perceptron. The method adopts the refractive-index field as the sole neural primitive and integrates multiresolution hash encoding, automatic-discrete gradient losses, and a three-dimensional mask to enable fast convergence and high-resolution, spatially coherent reconstructions. Tests on numerical combustion phantoms and real flame data demonstrate accurate recovery of both large-scale structures and fine-scale turbulence, strong robustness to noise, and clear advantages over frequency-encoding-based and voxel-based reconstruction methods.
title Neural Refractive Index Primitives for Flame Field Reconstruction Using Background-Oriented Schlieren
topic Fluid Dynamics
url https://arxiv.org/abs/2605.11454