cs.CV · 2605.00219 Copy arXiv ID · Apr 30, 2026 Save VkSplat: High-Performance 3DGS Training in Vulkan Compute Authors: Jingxiang Chen , Mohamed Ibrahim , Yang Liu
Organizations: Huawei Canada
Abstract We present VkSplat, a high-performance, cross-vendor 3D Gaussian Splatting (3DGS) training pipeline implemented fully in Vulkan compute, addressing performance and compatibility limitation of existing training pipelines. With various optimizations, we achieve 3.3 × 3.3\times 3.3 × 33 % 33\% 33% https://github.com/harry7557558/vksplat}{https://github.com/harry7557558/vksplat}
Explore similar work May 19, 2026 · Chonghao Zhong, Linfeng Shi, Hua Chen +4 3D Gaussian Splats Real3D-Ad Datasets
May 29, 2026 · Dawid Pająk, Martin Bisson, Rodolfo Lima Rasterization 3D Gaussian Splats
May 7, 2026 · Yongjae Lee, Jingxing Li, Abhay Kumar Yadav +2 3D Gaussian Splats Real3D-Ad Datasets
May 19, 2026 · cs.CV J/K move · Enter open · S save
Chonghao Zhong, Linfeng Shi, Hua Chen, Tiecheng Sun +3
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).