Research map / 3D vision / Gaussian splatting
Splat the Net: Radiance Fields with Splattable Neural Primitives
xilong zhou, Bao-Huy Nguyen, Loïc Magne, Vladislav Golyanik, Thomas Leimkuehler, Christian Theobalt
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 neural primitives instead of 3DGS under small memory budget | 3DGS · Synthetic NeRF, 0.1 MB | PSNR: 23.1 -> 24.7 | Improved |
| use neural primitives instead of 3DGS under 4 MB budget | 3DGS · Synthetic NeRF, 4.0 MB | PSNR: 29.6 -> 31.4 | Improved |
| use neural primitives instead of 3DGS with unlimited memory | 3DGS · Synthetic NeRF, unlimited | PSNR: 33.3 -> 33.4 | No clear effect |
| use neural primitives instead of 3DGS on real scenes | 3DGS · Mip-NeRF360 | memory (MB): 734 -> 93 | Improved |
| use neural primitives instead of 3DGS on real scenes | 3DGS · Mip-NeRF360 | FPS: 152 -> 115 | Hurt |
| use neural primitives instead of 3DGS | 3DGS · Deep Blending | PSNR: 29.41 -> 29.20 | Hurt |
| set frequency multiplier to 30 instead of 1 (ablation) | neural primitives · Mip-NeRF360 | PSNR: 26.32 -> 27.54 | Improved |
| increase hidden width from 8 to 16 (ablation) | neural primitives · Mip-NeRF360 | PSNR: 27.54 -> 27.62 | Improved |
| add geometry regularization against anisotropic primitives (ablation) | neural primitives · Mip-NeRF360 | PSNR | No clear effect |
| use random initialization instead of mesh initialization (ablation) | neural primitives · Synthetic NeRF | PSNR: 33.40 -> 33.36 | No clear effect |
| use AutoInt integration instead of closed-form antiderivative (ablation) | neural primitives · qualitative | multi-view consistency | Hurt |
Closest papers by meaning
- Compact 3D Gaussian Representation for Radiance Field (CVPR 2024 · similarity 0.92)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.91)
- HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis (NeurIPS 2025 · similarity 0.91)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.90)
- COLMAP-Free 3D Gaussian Splatting (CVPR 2024 · similarity 0.90)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · 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.
