Research map / 3D vision / Gaussian splatting
Lightweight Gradient-Aware Upscaling of 3D Gaussian Splatting Images
Simon Niedermayr, Christoph Neuhauser, Rüdiger Westermann
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 analytical Gaussian gradients (spline) instead of finite-difference bicubic upscaling during training | 3DGS with 4x upscaling · Tanks and Temples (4x scale) | PSNR: 20.28 -> 21.30 | Improved |
| use analytical Gaussian gradients (spline) instead of finite-difference bicubic upscaling during training | 3DGS with 4x upscaling · MipNeRF360 (4x scale) | PSNR: 26.60 -> 26.85 | Improved |
| use analytical Gaussian gradients (spline) instead of finite-difference bicubic upscaling during training | 3DGS with 4x upscaling · MipNeRF360 (4x scale) | LPIPS: 0.338 -> 0.322 | Improved |
| add upscaling into training (Spline Ours) rather than only at render time (Spline 3DGS) (ablation) | Spline upscaling · MipNeRF360 (4x scale) | PSNR: 25.71 -> 26.85 | Improved |
| add upscaling into training (Spline Ours) rather than only at render time (Spline 3DGS) (ablation) | Spline upscaling · Tanks and Temples (4x scale) | PSNR: 20.03 -> 21.30 | Improved |
| add upscaling into training (Spline Ours) rather than only at render time (Spline 3DGS) (ablation) | Spline upscaling · Deep Blending (4x scale) | PSNR: 29.14 -> 29.09 | Hurt |
| replace 3DGS full-resolution rendering with low-resolution rendering plus spline upscaling | 3DGS · MipNeRF360 (4x scale) | FPS: 15 -> 90 | Improved |
| replace 3DGS full-resolution rendering with low-resolution rendering plus spline upscaling | 3DGS · MipNeRF360 (4x scale) | PSNR: 26.35 -> 26.85 | Improved |
| replace 3DGS with Mip-Splatting | 3DGS · MipNeRF360 (4x scale) | FPS: 15 -> 10 | Hurt |
| spline upscaling at 8x scale versus 3DGS at 8x scale | 3DGS · MipNeRF360 (8x scale) | PSNR: 24.67 -> 25.95 | Improved |
| spline upscaling versus Mip-Splatting at 8x scale | Mip-Splatting · MipNeRF360 (8x scale) | PSNR: 26.37 -> 25.95 | Hurt |
| use spline upscaling instead of DLSS (no depth buffer) | 3DGS rendering · MipNeRF360 garden render time (3x upscale, 4096x2304) | speedup over 3DGS: x3.1 -> x4.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Mip-Splatting — Mip-Splatting: Alias-free 3D Gaussian Splatting (CVPR 2024) · better on 2 of 7 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
- Steepest Descent Density Control for Compact 3D Gaussian Splatting (CVPR 2025 · similarity 0.92)
- On Scaling Up 3D Gaussian Splatting Training (ICLR 2025 · similarity 0.91)
- Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis (CVPR 2024 · similarity 0.90)
- Sort-free Gaussian Splatting via Weighted Sum Rendering (ICLR 2025 · similarity 0.90)
- DashGaussian: Optimizing 3D Gaussian Splatting in 200 Seconds (CVPR 2025 · similarity 0.90)
- FlashGS: Efficient 3D Gaussian Splatting for Large-scale and High-resolution Rendering (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.
