GEMM-GS: Accelerating 3D Gaussian Splatting on Tensor Cores with GEMM-Compatible Blending
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917385398648832 |
|---|---|
| author | Li, Haomin Zhu, Bowen Liu, Fangxin Wang, Zongwu Liang, Xinran Jiang, Li Guan, Haibing |
| author_facet | Li, Haomin Zhu, Bowen Liu, Fangxin Wang, Zongwu Liang, Xinran Jiang, Li Guan, Haibing |
| contents | Neural Radiance Fields (NeRF) enables 3D scene reconstruction from several 2D images but incurs high rendering latency via its point-sampling design. 3D Gaussian Splatting (3DGS) improves on NeRF with explicit scene representation and an optimized pipeline yet still fails to meet practical real-time demands. Existing acceleration works overlook the evolving Tensor Cores of modern GPUs because 3DGS pipeline lacks General Matrix Multiplication (GEMM) operations. This paper proposes GEMM-GS, an acceleration approach utilizing tensor cores on GPUs via GEMM-friendly blending transformation. It equivalently reformulates the 3DGS blending process into a GEMM-compatible form to utilize Tensor Cores. A high-performance CUDA kernel is designed, integrating a three-stage double-buffered pipeline that overlaps computation and memory access. Extensive experiments show that GEMM-GS achieves $1.42\times$ speedup over vanilla 3DGS and provides an additional $1.47\times$ speedup on average when combining with existing acceleration approaches. Code is released at https://github.com/shieldforever/GEMM-GS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_02120 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GEMM-GS: Accelerating 3D Gaussian Splatting on Tensor Cores with GEMM-Compatible Blending Li, Haomin Zhu, Bowen Liu, Fangxin Wang, Zongwu Liang, Xinran Jiang, Li Guan, Haibing Hardware Architecture Graphics Neural Radiance Fields (NeRF) enables 3D scene reconstruction from several 2D images but incurs high rendering latency via its point-sampling design. 3D Gaussian Splatting (3DGS) improves on NeRF with explicit scene representation and an optimized pipeline yet still fails to meet practical real-time demands. Existing acceleration works overlook the evolving Tensor Cores of modern GPUs because 3DGS pipeline lacks General Matrix Multiplication (GEMM) operations. This paper proposes GEMM-GS, an acceleration approach utilizing tensor cores on GPUs via GEMM-friendly blending transformation. It equivalently reformulates the 3DGS blending process into a GEMM-compatible form to utilize Tensor Cores. A high-performance CUDA kernel is designed, integrating a three-stage double-buffered pipeline that overlaps computation and memory access. Extensive experiments show that GEMM-GS achieves $1.42\times$ speedup over vanilla 3DGS and provides an additional $1.47\times$ speedup on average when combining with existing acceleration approaches. Code is released at https://github.com/shieldforever/GEMM-GS. |
| title | GEMM-GS: Accelerating 3D Gaussian Splatting on Tensor Cores with GEMM-Compatible Blending |
| topic | Hardware Architecture Graphics |
| url | https://arxiv.org/abs/2604.02120 |