Research map / 3D vision / Gaussian splatting
PCGS: Deblurring 3D Gaussian Splatting with Patch Comparison
Yilong Li, Bo Pang, Zhongtao Wang, Mai Su, Yisong Chen, Chengwei Pan, Meng GAI, Fei Zhu, Guoping Wang
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 patch comparison (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.79 -> 28.01 | Improved |
| add patch comparison (ablation) | PCGS · Mip-NeRF360 | LPIPS: 0.211 -> 0.189 | Improved |
| add edge-based importance score for Gaussian selection (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.84 -> 28.01 | Improved |
| add edge-based importance score (ablation) | PCGS · Mip-NeRF360 | LPIPS: 0.203 -> 0.189 | Improved |
| use logarithmic budget instead of linear (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.87 -> 28.01 | Improved |
| use logarithmic budget instead of exponential (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.84 -> 28.01 | Improved |
| add Gaussian budget (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.88 -> 28.01 | Improved |
| add Gaussian budget (ablation) | PCGS · Mip-NeRF360 | primitives: 5310K -> 2500K | Improved |
| add Gaussian budget (ablation) | PCGS · Tanks&Temples | PSNR: 24.16 -> 24.40 | Improved |
| add Gaussian budget (ablation) | PCGS · Deep Blending | PSNR: 29.88 -> 30.04 | Improved |
| patch size 16 instead of 8 (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.97 -> 28.01 | No clear effect |
| patch size 16 instead of 32 (ablation) | PCGS · Mip-NeRF360 | PSNR: 27.90 -> 28.01 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Perceptual-GS — Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splatting (ICML 2025) · better on 3 of 3 shared results
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
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.93)
- EDGS: Eliminating Densification for Efficient Convergence of 3DGS (CVPR 2026 · similarity 0.92)
- Pushing Rendering Boundaries: Hard Gaussian Splatting (AAAI 2026 · similarity 0.92)
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- Steepest Descent Density Control for Compact 3D Gaussian Splatting (CVPR 2025 · 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.
