Research map / 3D vision / Gaussian splatting
RayGaussX: Accelerating Gaussian-Based Ray Marching for Real-Time and High-Quality Novel View Synthesis
Hugo Blanc, Jean-Emmanuel Deschaud, Alexis Paljic
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 new distance-corrected densification criterion (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | PSNR: 28.14 -> 28.38 | Improved |
| add new densification criterion (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | FPS: 1.5 -> 1.8 | Improved |
| add empty space skipping and adaptive sampling (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | FPS: 1.5 -> 5.9 | Improved |
| add empty space skipping and adaptive sampling (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | PSNR: 28.14 -> 28.14 | No clear effect |
| add Z-order reindexing and ray coherence (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | FPS: 1.5 -> 3.6 | Improved |
| add isotropic scale loss (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | FPS: 1.5 -> 10.1 | Improved |
| add isotropic scale loss (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | PSNR: 28.14 -> 28.12 | Hurt |
| add isotropic scale loss (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | Train time: 297min -> 133min | Improved |
| combine all four contributions (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | FPS: 1.5 -> 27.4 | Improved |
| combine all four contributions (ablation) | optimized RayGauss · Mip-NeRF360 garden (ablation) | Train time: 297min -> 84min | Improved |
| replace RayGauss with RayGaussX | RayGauss · Tanks&Temples | PSNR: 23.20 -> 23.76 | Improved |
| replace RayGauss with RayGaussX | RayGauss · Mip-NeRF360 (9 scenes) | FPS: 0.5 -> 24.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- Zip-NeRF — Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields (ICCV 2023) · better on 1 of 3 shared results
- Mip-Splatting — Mip-Splatting: Alias-free 3D Gaussian Splatting (CVPR 2024) · better on 1 of 1 shared result
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
- Hardware-Rasterized Ray-Based Gaussian Splatting (CVPR 2025 · similarity 0.87)
- Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering (CVPR 2025 · similarity 0.87)
- VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction (CVPR 2024 · similarity 0.86)
- Gaussian Herding across Pens: An Optimal Transport Perspective on Global Gaussian Reduction for 3DGS (NeurIPS 2025 · similarity 0.86)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.86)
- MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks (CVPR 2025 · similarity 0.86)
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.
