TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization

Hugging Face Daily Papers Papers
3d-gaussian-splatting out-of-core training scalable single-gpu memory-optimization

Summary

TideGS introduces an out-of-core training framework that enables 3D Gaussian Splatting with over one billion primitives on a single GPU by managing parameters across SSD-CPU-GPU hierarchy via block-virtualization, asynchronous pipeline, and differential streaming techniques.

Training 3D Gaussian Splatting (3DGS) at billion-primitive scale is fundamentally memory-bound: each Gaussian primitive carries a large attribute vector, and the aggregate parameter table quickly exceeds GPU capacity, limiting prior systems to tens of millions of Gaussians on commodity single-GPU hardware. We observe that 3DGS training is inherently sparse and trajectory-conditioned: each iteration activates only the Gaussians visible from the current camera batch, so GPU memory can serve as a working-set cache rather than a persistent parameter store. Building on this insight, we introduce TideGS, an out-of-core training framework that manages parameters across an SSD-CPU-GPU hierarchy via three synergistic techniques: block-virtualized geometry for SSD-aligned spatial locality, a hierarchical asynchronous pipeline to overlap I/O with computation, and trajectory-adaptive differential streaming that transfers only incremental working-set deltas between iterations. Experiments show that TideGS enables training with over one billion Gaussians on a single 24 GB GPU while achieving the best reconstruction quality among evaluated single-GPU baselines on large-scale scenes, scaling beyond prior out-of-core baselines (e.g., approximately 100M Gaussians) and standard in-memory training (e.g., approximately 11M Gaussians).
Original Article
View Cached Full Text

Cached at: 05/20/26, 10:37 AM

Paper page - TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization

Source: https://huggingface.co/papers/2605.20150

Abstract

TideGS enables training 3D Gaussian Splatting with over one billion primitives on a single GPU by managing parameters across SSD-CPU-GPU hierarchy through block-virtualization, asynchronous pipeline, and differential streaming techniques.

Training3D Gaussian Splatting(3DGS) at billion-primitive scale is fundamentally memory-bound: each Gaussian primitive carries a large attribute vector, and the aggregate parameter table quickly exceeds GPU capacity, limiting prior systems to tens of millions of Gaussians on commodity single-GPU hardware. We observe that 3DGS training is inherently sparse and trajectory-conditioned: each iteration activates only the Gaussians visible from the current camera batch, so GPU memory can serve as aworking-set cacherather than a persistent parameter store. Building on this insight, we introduce TideGS, anout-of-core trainingframework that manages parameters across anSSD-CPU-GPU hierarchyvia three synergistic techniques:block-virtualized geometryfor SSD-aligned spatial locality, ahierarchical asynchronous pipelineto overlap I/O with computation, andtrajectory-adaptive differential streamingthat transfers only incremental working-set deltas between iterations. Experiments show that TideGS enables training with over one billion Gaussians on a single 24 GB GPU while achieving the best reconstruction quality among evaluated single-GPU baselines on large-scale scenes, scaling beyond prior out-of-core baselines (e.g., approximately 100M Gaussians) and standard in-memory training (e.g., approximately 11M Gaussians).

View arXiv pageView PDFProject pageGitHub12Add to collection

Get this paper in your agent:

hf papers read 2605\.20150

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.20150 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.20150 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.20150 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Gaussian Point Splatting

Hacker News Top

Researchers propose Gaussian Point Splatting, a stochastic rendering method using pixel-sized opaque points and 64-bit GPU atomics that scales to hundreds of millions of Gaussians in real time. The method, accepted at SIGGRAPH 2026, employs hierarchical culling and parallel programming primitives to achieve even workload distribution with only minor noise differences compared to original Gaussian splatting.

GlobalSplat: Efficient Feed-Forward 3D Gaussian Splatting via Global Scene Tokens

Hugging Face Daily Papers

GlobalSplat introduces an efficient feed-forward framework for 3D Gaussian splatting that achieves compact and consistent scene reconstruction using global scene tokens, reducing computational overhead and inference time to under 78ms. The method uses a coarse-to-fine training approach to prevent representation bloat while maintaining competitive novel-view synthesis performance with significantly fewer Gaussians (16K) compared to dense baselines.

Better Gaussian Splatting in Julia

Hacker News Top

GaussianSplatting.jl 2.0 release brings multi-GPU backend support via KernelAbstractions.jl, a multithreaded UI, MCMC densification strategy, and depth/geometry supervision for better 3D reconstruction.