Research map / 3D vision / Gaussian splatting
Mobile-GS: Real-time Gaussian Splatting for Mobile Devices
Xiaobiao Du, Yida Wang, Kun Zhan, Xin Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use depth-aware order-independent rendering instead of alpha blending (ablation) | Mobile-GS · Mip-NeRF360 | FPS: 684 -> 1125 | Improved |
| use depth-aware order-independent rendering instead of alpha blending (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.26 -> 27.12 | Hurt |
| add neural view-dependent enhancement (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 26.68 -> 27.12 | Improved |
| add neural vector quantization (ablation) | Mobile-GS · Mip-NeRF360 | Storage: 121 MB -> 4.6 MB | Improved |
| add neural vector quantization (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.33 -> 27.12 | Hurt |
| use first-order SH distillation instead of 3rd-order SH (ablation) | Mobile-GS · Mip-NeRF360 | FPS: 841 -> 1125 | Improved |
| use first-order SH instead of 0th-order (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.04 -> 27.12 | Improved |
| add depth term to blending weight (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.03 -> 27.12 | Improved |
| add scale term to blending weight (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.08 -> 27.12 | Improved |
| prune using opacity and scale jointly vs opacity only (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 26.84 -> 27.12 | Improved |
| increase pruning threshold from 0.2 to 0.6 (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.12 -> 25.85 | Hurt |
| increase codebook size from 2^10 to 2^12 (ablation) | Mobile-GS · Mip-NeRF360 | PSNR: 27.12 -> 27.15 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports Mobile-GS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Monte Carlo Energy Aggregation for Mobile 3D Gaussian Splatting (ECCV 2026 · better on 17 of 20 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.93)
- Sort-free Gaussian Splatting via Weighted Sum Rendering (ICLR 2025 · similarity 0.92)
- LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering (NeurIPS 2025 · similarity 0.91)
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- LOD-GS: Achieving Levels of Detail using Scalable Gaussian Soup (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
