Research map / 3D vision / Gaussian splatting
GS2Mesh: Surface Reconstruction from Gaussian Splatting via Novel Stereo Views
Yaniv Wolf, Amit Bracha, Ron Kimmel
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 |
|---|---|---|---|
| extract depth from novel stereo views with pre-trained stereo model instead of Gaussian locations | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.74 -> 0.68 vs GOF; 1.33 -> 0.68 vs SuGaR | Improved |
| use DLNR stereo model instead of RAFT (ablation) | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.73 -> 0.68 | Improved |
| use stereo baseline 7% of scene radius instead of 3.5% or 10.5% (ablation) | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.79 / 0.73 -> 0.68 | Improved |
| add occlusion mask (ablation) | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.77 -> 0.68 | Improved |
| train 3DGS for 30000 iterations instead of 7000 (ablation) | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.70 -> 0.68 | No clear effect |
| train 3DGS for 60000 iterations instead of 30000 (ablation) | GS2Mesh · DTU (15 scans) | Chamfer distance (mean): 0.68 -> 0.69 | No clear effect |
| use novel stereo views plus stereo model instead of deep MVS on original images (ablation) | GS2Mesh · MobileBrick test | Chamfer distance (mm): 9.11 -> 4.94 | Improved |
| feed 3DGS-rendered images to MVSFormer instead of original images (ablation) | MVSFormer · MobileBrick test | Chamfer distance (mm): 9.11 -> 7.10 | Improved |
| use stereo pipeline instead of MVSFormer on rendered images (ablation) | GS2Mesh · MobileBrick test (2.5mm radius) | Accuracy: 80.16 -> 68.77 | Hurt |
| apply GS2Mesh versus Neuralangelo on Tanks and Temples | GS2Mesh · Tanks and Temples (4 scenes) | mean F1: 0.61 -> 0.37 | Hurt |
Later papers that compared with this method
1 later accepted paper reports GS2Mesh as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization (ICCV 2025 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · better on 1 of 2 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
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.92)
- GS-2DGS: Geometrically Supervised 2DGS for Reflective Object Reconstruction (CVPR 2025 · similarity 0.92)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.92)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.92)
- SfM-Free 3D Gaussian Splatting via Hierarchical Training (CVPR 2025 · similarity 0.90)
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (CVPR 2024 · 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.
