Research map / 3D vision / Gaussian splatting
Eulerian Gaussian Splatting using Hashed Probability Pyramids
Mia Gaia Polansky, George Kopanas, Stephan Garbin, Todd Zickler, Dor Verbin
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 |
|---|---|---|---|
| add opacity regularization (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), post-refinement | PSNR: 26.74 -> 29.52 | Improved |
| add opacity regularization (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), pre-refinement | PSNR: 26.32 -> 29.29 | Improved |
| add scaling regularization (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), post-refinement | PSNR: 29.48 -> 29.52 | Improved |
| add sample rounding (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), pre-refinement | PSNR: 28.85 -> 29.29 | Improved |
| add sample rounding (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), post-refinement | PSNR: 29.44 -> 29.52 | Improved |
| add defensive sampling (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), post-refinement | PSNR: 29.45 -> 29.52 | Improved |
| use control-variate gradient estimator instead of autodiff (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), pre-refinement | PSNR: 22.35 -> 29.29 | Improved |
| use control-variate gradient estimator instead of autodiff (ablation) | EGS · Mip-NeRF 360 (6-scene ablation), post-refinement | PSNR: 23.69 -> 29.52 | Improved |
| add 5K refinement iterations after probabilistic training | EGS · Mip-NeRF 360 (9 scenes, average) | PSNR: 28.10 -> 28.35 | Improved |
| random init instead of COLMAP init for 3DGS-MCMC | 3DGS-MCMC · Mip-NeRF 360 (9 scenes, average) | PSNR: 28.38 -> 28.26 | Hurt |
| EGS vs 3DGS-MCMC under random init | EGS refined · Mip-NeRF 360 (9 scenes, average) | PSNR: 28.26 -> 28.35 | Improved |
| EGS vs 3DGS-MCMC under random init | EGS refined · Deep Blending (2 scenes, average) | PSNR: 29.72 -> 29.85 | Improved |
Closest papers by meaning
- 3D Gaussian Splatting as Markov Chain Monte Carlo (NeurIPS 2024 · similarity 0.92)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.92)
- Fully Explicit Dynamic Gaussian Splatting (NeurIPS 2024 · similarity 0.91)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- AdpSplit: Error-Driven Adaptive Splitting for Faster Geometry Discovery in 3D Gaussian Splatting (NeurIPS 2026 · similarity 0.91)
- ReorgGS: Equivalent Distribution Reorganization for 3D Gaussian Splatting (NeurIPS 2026 · similarity 0.91)
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.
