Voyager: Real-Time Splatting City-Scale Gaussians on Resource-Constrained Devices

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Liu, Zheng, Zhu, He, Li, Xinyang, Wang, Yirun, Shi, Yujiao, Gan, Yiming, Li, Wei, Leng, Jingwen, Guo, Minyi, Feng, Yu
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909807920807936
author Liu, Zheng
Zhu, He
Li, Xinyang
Wang, Yirun
Shi, Yujiao
Gan, Yiming
Li, Wei
Leng, Jingwen
Guo, Minyi
Feng, Yu
author_facet Liu, Zheng
Zhu, He
Li, Xinyang
Wang, Yirun
Shi, Yujiao
Gan, Yiming
Li, Wei
Leng, Jingwen
Guo, Minyi
Feng, Yu
contents 3D Gaussian splatting (3DGS) is an emerging technique for photorealistic 3D scene rendering. However, rendering city-scale 3DGS scenes on resource-constrained mobile devices in real-time remains a significant challenge due to two compute-intensive stages: level-of-detail (LoD) search and rasterization. In this paper, we propose Voyager, an effective solution to accelerate city-scale 3DGS rendering on mobile devices. Our key insight is that, under normal user motion, the number of newly visible Gaussians within the view frustum remains roughly constant. Leveraging this temporal correlation, we propose a temporal-aware LoD search to identify the necessary Gaussians for the remaining rendering stages. For the remaining rendering process, we accelerate the bottleneck stage, rasterization, via preemptive $α$-filtering. With all optimizations above, our system can deliver low-latency, city-scale 3DGS rendering on mobile devices. Compared to existing solutions, Voyager achieves up to 6.6$\times$ speedup and 85\% energy savings with superior rendering quality.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02774
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Voyager: Real-Time Splatting City-Scale Gaussians on Resource-Constrained Devices
Liu, Zheng
Zhu, He
Li, Xinyang
Wang, Yirun
Shi, Yujiao
Gan, Yiming
Li, Wei
Leng, Jingwen
Guo, Minyi
Feng, Yu
Graphics
3D Gaussian splatting (3DGS) is an emerging technique for photorealistic 3D scene rendering. However, rendering city-scale 3DGS scenes on resource-constrained mobile devices in real-time remains a significant challenge due to two compute-intensive stages: level-of-detail (LoD) search and rasterization. In this paper, we propose Voyager, an effective solution to accelerate city-scale 3DGS rendering on mobile devices. Our key insight is that, under normal user motion, the number of newly visible Gaussians within the view frustum remains roughly constant. Leveraging this temporal correlation, we propose a temporal-aware LoD search to identify the necessary Gaussians for the remaining rendering stages. For the remaining rendering process, we accelerate the bottleneck stage, rasterization, via preemptive $α$-filtering. With all optimizations above, our system can deliver low-latency, city-scale 3DGS rendering on mobile devices. Compared to existing solutions, Voyager achieves up to 6.6$\times$ speedup and 85\% energy savings with superior rendering quality.
title Voyager: Real-Time Splatting City-Scale Gaussians on Resource-Constrained Devices
topic Graphics
url https://arxiv.org/abs/2506.02774