A Scalable PyTorch Abstraction for Multi-GPU Gaussian Splatting

• URL: http://arxiv.org/abs/2606.11390v1
• Date: Tue, 09 Jun 2026 19:15:39 GMT
• Title: A Scalable PyTorch Abstraction for Multi-GPU Gaussian Splatting
• Abstract summary: We present a multi-GPU approach to Gaussian splatting that scales reconstruction to higher resolutions and larger scenes.<n>We propose a PyTorch backend that distributes the parameters and splatting operators across GPUs via unified memory and NVLink.<n>We demonstrate city-scale reconstructions with street-level detail consisting of over 1 billion Gaussian splats, more than 25 times as many as the current state of the art.
• Score: 44.78530292755235
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Gaussian splatting methods have become increasingly popular for neural reconstruction of the real world. However, they are often limited in scale and resolution due to compute and memory constraints. We present a multi-GPU Gaussian splatting approach that scales reconstruction to higher resolutions and larger scenes while abstracting away the code complexity typically associated with distributing a model. To accomplish this, we propose a PyTorch backend that distributes the Gaussian parameters and splatting operators across GPUs via CUDA unified memory and NVLink. Because distribution occurs at the operator level, the model code requires no explicit cross-device communication. More broadly, the backend exposes multiple GPUs as an aggregate PyTorch device and supports other PyTorch operators. We demonstrate city-scale reconstructions with street-level detail consisting of over 1 billion Gaussian splats, more than 25 times as many as the current state of the art.

Related papers

• TorchMorph: CUDA-accelerated Morphological Transforms [4.910415354181461]
  GPU vision libraries built on PyTorch cover a narrow subset of operators.<n>TorchMorph exposes operators covering binary morphology, greyscale morphology and exact and approximate distance transforms.<n> Batched execution reaches up to 1.1e3 times the throughput of scipy.ndimage on greyscale morphology.
  arXiv  Detail & Related papers  (2026-08-25T15:46:45Z)
• StochasticSplats: Stochastic Rasterization for Sorting-Free 3D Gaussian Splatting [28.48846639700183]
  3D Gaussian splatting (3DGS) is a popular radiance field method, with many application-specific extensions.<n>Most variants rely on the same core algorithm: depth-sorting of Gaussian splats then rendering in primitive order.<n>We address the above limitations by combining 3D splatting with rendering.<n>Our method renders more than four times faster than sortedization.
  arXiv  Detail & Related papers  (2025-03-31T17:46:18Z)
• Occam's LGS: An Efficient Approach for Language Gaussian Splatting [57.00354758206751]
  We show that the complicated pipelines for language 3D Gaussian Splatting are simply unnecessary.<n>We apply Occam's razor to the task at hand, leading to a highly efficient weighted multi-view feature aggregation technique.
  arXiv  Detail & Related papers  (2024-12-02T18:50:37Z)
• SmileSplat: Generalizable Gaussian Splats for Unconstrained Sparse Images [91.28365943547703]
  A novel generalizable Gaussian Splatting method, SmileSplat, is proposed to reconstruct pixel-aligned Gaussian surfels for diverse scenarios.<n>The proposed method achieves state-of-the-art performance in various 3D vision tasks.
  arXiv  Detail & Related papers  (2024-11-27T05:52:28Z)
• PixelGaussian: Generalizable 3D Gaussian Reconstruction from Arbitrary Views [116.10577967146762]
  PixelGaussian is an efficient framework for learning generalizable 3D Gaussian reconstruction from arbitrary views. Our method achieves state-of-the-art performance with good generalization to various numbers of views.
  arXiv  Detail & Related papers  (2024-10-24T17:59:58Z)
• HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstruction [46.269350101349715]
  HiSplat is a novel framework for generalizable 3D Gaussian Splatting. It generates hierarchical 3D Gaussians via a coarse-to-fine strategy. It significantly enhances reconstruction quality and cross-dataset generalization.
  arXiv  Detail & Related papers  (2024-10-08T17:59:32Z)
• On Scaling Up 3D Gaussian Splatting Training [31.161086345038424]
  3DGS is increasingly popular for 3D reconstruction due to its superior visual quality and rendering speed.<n>Currently, 3DGS training occurs on a single GPU, limiting its ability to handle high-resolution and large-scale 3D reconstruction tasks.<n>We introduce Grendel, a distributed system designed to partition 3DGS parameters and parallelize across multiple GPU.
  arXiv  Detail & Related papers  (2024-06-26T17:59:28Z)
• PyGS: Large-scale Scene Representation with Pyramidal 3D Gaussian Splatting [15.378596308239667]
  We present Pyramidal 3D Gaussian Splatting (PyGS) with NeRF Initialization. We group these pyramidal Gaussians into clusters and use a compact weighting network to dynamically determine the influence of each pyramid level of each cluster. Our method achieves a significant performance leap across multiple large-scale datasets and attains a rendering time that is over 400 times faster than current state-of-the-art approaches.
  arXiv  Detail & Related papers  (2024-05-27T04:52:21Z)
• GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting [49.32327147931905]
  We propose GS-LRM, a scalable large reconstruction model that can predict high-quality 3D Gaussians from 2-4 posed sparse images in 0.23 seconds on single A100 GPU. Our model features a very simple transformer-based architecture; we patchify input posed images, pass the primitive multi-view image tokens through a sequence of transformer blocks, and decode final per-pixel Gaussian parameters directly from these tokens for differentiable rendering.
  arXiv  Detail & Related papers  (2024-04-30T16:47:46Z)
• Adaptive Elastic Training for Sparse Deep Learning on Heterogeneous Multi-GPU Servers [65.60007071024629]
  We show that Adaptive SGD outperforms four state-of-the-art solutions in time-to-accuracy. We show experimentally that Adaptive SGD outperforms four state-of-the-art solutions in time-to-accuracy.
  arXiv  Detail & Related papers  (2021-10-13T20:58:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.