WebSplatter: Enabling Cross-Device Efficient Gaussian Splatting in Web Browsers via WebGPU

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Han, Yudong, Xu, Chao, Ye, Xiaodan, Bi, Weichen, Dong, Zilong, Ma, Yun
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912871809548288
author Han, Yudong
Xu, Chao
Ye, Xiaodan
Bi, Weichen
Dong, Zilong
Ma, Yun
author_facet Han, Yudong
Xu, Chao
Ye, Xiaodan
Bi, Weichen
Dong, Zilong
Ma, Yun
contents We present WebSplatter, an end-to-end GPU rendering pipeline for the heterogeneous web ecosystem. Unlike naive ports, WebSplatter introduces a wait-free hierarchical radix sort that circumvents the lack of global atomics in WebGPU, ensuring deterministic execution across diverse hardware. Furthermore, we propose an opacity-aware geometry culling stage that dynamically prunes splats before rasterization, significantly reducing overdraw and peak memory footprint. Evaluation demonstrates that WebSplatter consistently achieves 1.2$\times$ to 4.5$\times$ speedups over state-of-the-art web viewers.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03207
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WebSplatter: Enabling Cross-Device Efficient Gaussian Splatting in Web Browsers via WebGPU
Han, Yudong
Xu, Chao
Ye, Xiaodan
Bi, Weichen
Dong, Zilong
Ma, Yun
Graphics
Computer Vision and Pattern Recognition
Performance
I.3.7; I.3.1
We present WebSplatter, an end-to-end GPU rendering pipeline for the heterogeneous web ecosystem. Unlike naive ports, WebSplatter introduces a wait-free hierarchical radix sort that circumvents the lack of global atomics in WebGPU, ensuring deterministic execution across diverse hardware. Furthermore, we propose an opacity-aware geometry culling stage that dynamically prunes splats before rasterization, significantly reducing overdraw and peak memory footprint. Evaluation demonstrates that WebSplatter consistently achieves 1.2$\times$ to 4.5$\times$ speedups over state-of-the-art web viewers.
title WebSplatter: Enabling Cross-Device Efficient Gaussian Splatting in Web Browsers via WebGPU
topic Graphics
Computer Vision and Pattern Recognition
Performance
I.3.7; I.3.1
url https://arxiv.org/abs/2602.03207