Research map / 3D vision / Gaussian splatting
ResGS: Residual Densification of 3D Gaussian for Efficient Detail Recovery
Yanzhe Lyu, Kai Cheng, Xin Kang, Xuejin Chen
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 residual split to base pipeline (ablation) | Base · Mip-NeRF360 (ablation) | PSNR: 27.41 -> 27.66 | Improved |
| add image pyramid supervision to base (ablation) | Base · Mip-NeRF360 (ablation) | PSNR: 27.41 -> 27.54 | Improved |
| add image pyramid supervision to base (ablation) | Base · Deep Blending (ablation) | PSNR: 29.60 -> 29.05 | Hurt |
| add image pyramid on top of residual split (ablation) | Base + RS · Mip-NeRF360 (ablation) | PSNR: 27.66 -> 27.88 | Improved |
| add image pyramid on top of residual split (ablation) | Base + RS · Deep Blending (ablation) | PSNR: 29.68 -> 29.82 | Improved |
| add varying gradient threshold (ablation) | Base + RS + IP · Mip-NeRF360 (ablation) | PSNR: 27.88 -> 28.00 | Improved |
| add varying gradient threshold (ablation) | Base + RS + IP · Deep Blending (ablation) | PSNR: 29.82 -> 29.91 | Improved |
| replace split and clone with residual split | AbsGS · Mip-NeRF360 (9 scenes) | PSNR: 27.49 -> 27.71 | Improved |
| replace split and clone with residual split | 3D-GS · Mip-NeRF360 (9 scenes) | PSNR: 27.21 -> 27.44 | Improved |
| replace split and clone with residual split | Pixel-GS · Deep Blending (2 scenes) | PSNR: 28.91 -> 29.75 | Improved |
| replace split and clone with residual split | 3D-GS · Mip-NeRF360 (9 scenes) | memory: 734MB -> 586MB | Improved |
| vary beta in [0.05,1.0) and lambda_s in [1.05,2.0] (ablation) | ResGS · Tanks&Temples | LPIPS: 0.006 LPIPS difference between best and worst | No clear effect |
Closest papers by meaning
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- FlashGS: Efficient 3D Gaussian Splatting for Large-scale and High-resolution Rendering (CVPR 2025 · similarity 0.90)
- Steepest Descent Density Control for Compact 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (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.
