Research map / 3D vision / Gaussian splatting
MeshSplat: Generalizable Sparse-View Surface Reconstruction via Gaussian Splatting
Hanzhi Chang, Ruijie Zhu, Wenjie Chang, Mulin Yu, Yanzhe Liang, Jiahao Lu, Zhuoyuan Li, Tianzhu Zhang
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 |
|---|---|---|---|
| use 2DGS instead of 3DGS as the feed-forward representation (ablation) | MeshSplat · Scannet | Chamfer distance: 0.3748 -> 0.2948 | Improved |
| add Weighted Chamfer Distance loss and normal prediction network to 2DGS (ablation) | MeshSplat · Scannet | Chamfer distance: 0.2948 -> 0.2606 | Improved |
| add only one of WCD loss or normal prediction network to 2DGS (ablation) | MeshSplat · Scannet | Chamfer distance: 0.2948 -> 0.2769 and 0.2642 (intermediate rows) | Improved |
| use 2DGS without regularization instead of 3DGS (ablation) | Gaussian normals · Rendered normal maps (qualitative) | normal quality | Hurt |
| add normal prediction network with NLL loss (ablation) | 2DGS · Rendered normal maps (qualitative) | normal accuracy | Improved |
| replace MVSplat with MeshSplat for depth prediction | MeshSplat · Depth prediction | AbsRel: 0.1692 -> 0.0910 | Improved |
| MeshSplat vs MVSplat for normal prediction | MeshSplat · Normal prediction | Mean angular error: 57.16 -> 33.84 | Improved |
| zero-shot transfer from Re10K to Replica | MeshSplat · Re10K to Replica | F1: 0.0564 -> 0.0809 | Improved |
| MeshSplat vs MVSplat on Re10K | MeshSplat · Re10K | Chamfer distance: 0.4015 -> 0.3566 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MVSplat — MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images (ECCV 2024) · better on 10 of 10 shared results
- pixelSplat — pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024) · 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
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025 · similarity 0.94)
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.94)
- TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers (AAAI 2025 · similarity 0.91)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.91)
- SurfelSplat: Learning Efficient and Generalizable Gaussian Surfel Representations for Sparse-View Surface Reconstruction (NeurIPS 2025 · similarity 0.91)
- 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.
