Research map / 3D vision / Gaussian splatting
Quadratic Gaussian Splatting: High Quality Surface Reconstruction with Second-order Geometric Primitives
Ziyu Zhang, Binbin Huang, Hanqing Jiang, Liyang Zhou, Xiaojun Xiang, Shuhan Shen
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 per-pixel resorting (ablation) | QGS · TNT | mean F1: 0.48 -> 0.50 | Improved |
| add curvature-guided normal consistency (ablation) | QGS · TNT | mean F1: 0.49 -> 0.50 | Improved |
| add multi-view regularization (ablation) | QGS · TNT | mean F1: 0.45 -> 0.50 | Improved |
| add multi-view regularization (ablation) | QGS · DTU | Chamfer distance (mm): 0.71 -> 0.54 | Improved |
| use quadric with free z-scale instead of disk (s3 fixed) (ablation) | QGS w/o MV · TNT | mean F1: 0.37 -> 0.45 | Improved |
| use geodesic distance instead of Euclidean distance (ablation) | QGS w/o MV · TNT | mean F1: 0.38 -> 0.45 | Improved |
| quadric surfels instead of 2DGS disks (no MV) | surface reconstruction · TNT | mean F1: 0.33 -> 0.45 | Improved |
| quadric surfels instead of 2DGS disks (no MV) | surface reconstruction · DTU | Chamfer distance (mm): 0.80 -> 0.71 | Improved |
| QGS compared with GOF without MV | surface reconstruction · TNT | mean F1: 0.46 vs 0.45 | No clear effect |
| use QGS instead of 2DGS | rendering · Mip-NeRF 360 (outdoor) | PSNR: 24.33 -> 24.32 | No clear effect |
| use QGS instead of GOF | rendering · Mip-NeRF 360 (outdoor) | PSNR: 24.76 -> 24.32 | Hurt |
| add tighter bounding box and memory optimization (ablation) | QGS · Mip-NeRF 360 | FPS: 7.92 -> 14.15 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- GSDF — GSDF: 3DGS Meets SDF for Improved Neural Rendering and Reconstruction (NeurIPS 2024) · better on 1 of 1 shared result
- Geo-NeuS — Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction (NeurIPS 2022) · better on 1 of 1 shared result
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · 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
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025 · similarity 0.90)
- Geometry Field Splatting with Gaussian Surfels (CVPR 2025 · similarity 0.90)
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.89)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.89)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (CVPR 2025 · similarity 0.89)
- Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering (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.
