Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.05216 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908521398796288 |
|---|---|
| author | Han, Mengjiao Sewell, Andres Insley, Joseph Knowles, Janet Mateevitsi, Victor A. Papka, Michael E. Petruzza, Steve Rizzi, Silvio |
| author_facet | Han, Mengjiao Sewell, Andres Insley, Joseph Knowles, Janet Mateevitsi, Victor A. Papka, Michael E. Petruzza, Steve Rizzi, Silvio |
| contents | We present a multi-GPU extension of the 3D Gaussian Splatting (3D-GS) pipeline for scientific visualization. Building on previous work that demonstrated high-fidelity isosurface reconstruction using Gaussian primitives, we incorporate a multi-GPU training backend adapted from Grendel-GS to enable scalable processing of large datasets. By distributing optimization across GPUs, our method improves training throughput and supports high-resolution reconstructions that exceed single-GPU capacity. In our experiments, the system achieves a 5.6X speedup on the Kingsnake dataset (4M Gaussians) using four GPUs compared to a single-GPU baseline, and successfully trains the Miranda dataset (18M Gaussians) that is an infeasible task on a single A100 GPU. This work lays the groundwork for integrating 3D-GS into HPC-based scientific workflows, enabling real-time post hoc and in situ visualization of complex simulations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_05216 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Toward Distributed 3D Gaussian Splatting for High-Resolution Isosurface Visualization Han, Mengjiao Sewell, Andres Insley, Joseph Knowles, Janet Mateevitsi, Victor A. Papka, Michael E. Petruzza, Steve Rizzi, Silvio Distributed, Parallel, and Cluster Computing We present a multi-GPU extension of the 3D Gaussian Splatting (3D-GS) pipeline for scientific visualization. Building on previous work that demonstrated high-fidelity isosurface reconstruction using Gaussian primitives, we incorporate a multi-GPU training backend adapted from Grendel-GS to enable scalable processing of large datasets. By distributing optimization across GPUs, our method improves training throughput and supports high-resolution reconstructions that exceed single-GPU capacity. In our experiments, the system achieves a 5.6X speedup on the Kingsnake dataset (4M Gaussians) using four GPUs compared to a single-GPU baseline, and successfully trains the Miranda dataset (18M Gaussians) that is an infeasible task on a single A100 GPU. This work lays the groundwork for integrating 3D-GS into HPC-based scientific workflows, enabling real-time post hoc and in situ visualization of complex simulations. |
| title | Toward Distributed 3D Gaussian Splatting for High-Resolution Isosurface Visualization |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2509.05216 |