Research map / 3D vision / Gaussian splatting
Neural Gabor Splatting: Enhanced Gaussian Splatting with Neural Gabor for High-frequency Surface Reconstruction
Haato Watanabe, Nobuyuki Umetani
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 the neural Gabor primitive and frequency-aware densification versus 2DGS under equal data budget | 2DGS · High-Frequency | PSNR: 23.91 -> 26.49 | Improved |
| replace 3DGS single-color primitives with neural Gabor primitives under equal data budget | 3DGS · High-Frequency | PSNR: 23.97 -> 26.49 | Improved |
| neural Gabor primitives versus 3DGS on Mip-NeRF360 at same budget | 3DGS · Mip-NeRF360 (7 scenes) | PSNR: 27.23 -> 26.98 | Hurt |
| neural Gabor primitives versus 3DGS on Mip-NeRF360 | 3DGS · Mip-NeRF360 (7 scenes) | LPIPS: 0.2931 -> 0.2521 | Improved |
| neural Gabor primitives versus 2DGS on DTU | 2DGS · DTU (4 scans) | LPIPS: 0.4171 -> 0.3705 | Improved |
| neural Gabor primitives versus 2DGS on Tanks and Temples | 2DGS · Tanks and Temples (2 scenes) | PSNR: 21.12 -> 21.93 | Improved |
| neural MLP primitive versus 3D Gabor splatting with frequency-aware densification | 3D Gabor splatting · High-Frequency | PSNR: 25.28 -> 26.02 | Improved |
| neural MLP primitive versus 3D Gabor splatting on Mip-NeRF360 | 3D Gabor splatting · Mip-NeRF360 (7 scenes) | PSNR: 26.76 -> 26.67 | No clear effect |
| add view-dependent modeling to the neural Gabor primitive (ablation) | Neural Gabor splatting · High-Frequency | PSNR: 26.02 -> 26.49 | Improved |
| add view-dependent modeling to the neural Gabor primitive (ablation) | Neural Gabor splatting · DTU (4 scans) | PSNR: 29.50 -> 29.15 | Hurt |
| neural Gabor splatting versus NTS under same memory footprint | NTS · High-Frequency (4-scene avg, NEST/NTS comparison) | PSNR: 23.48 -> 26.49 | Improved |
| neural Gabor splatting versus NTS on Bonsai | NTS · Mip-NeRF360 (Bonsai only) | PSNR: 29.49 -> 29.18 | Hurt |
Closest papers by meaning
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.89)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.89)
- FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Training (NeurIPS 2024 · similarity 0.89)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (AAAI 2025 · similarity 0.89)
- SGSST: Scaling Gaussian Splatting Style Transfer (CVPR 2025 · similarity 0.89)
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.
