Research map / 3D vision / Face and shape
Mesh Splatting for End-to-end Multiview Surface Reconstruction
Ruiqi Zhang, JiachengWU, Jie 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 multi-layer mesh softening (volumetric supervision) over shading-only (ablation) | Mesh Splatting · DTU | Chamfer distance (cm): 0.73 -> 0.62 | Improved |
| add multi-layer mesh softening over shading-only (ablation) | Mesh Splatting · BlendedMVS | Chamfer distance (cm): 1.94 -> 1.71 | Improved |
| replace iterative mesh rasterization with mesh splatting (ablation) | softened mesh renderer · DTU scan122 (1/4 scale) | GPU memory (GB): 8 -> 2 | Improved |
| replace iterative mesh rasterization with mesh splatting (ablation) | softened mesh renderer · DTU scan122 (1/2 scale) | training time (min): 90 -> 15 | Improved |
| replace iterative mesh rasterization with mesh splatting (ablation) | softened mesh renderer · DTU scan122 (full scale) | feasibility: OOM -> 13 GB, 22 min | Improved |
| use Gaussian Splatting instead of mesh splatting (ablation) | renderer · DTU scan122 (full scale) | training time (min): 22 -> 13 | Improved |
| add DMTet early stage (hybrid topology) to remeshing (ablation) | Full model · DTU | Chamfer distance: 3.79 -> 1.57 | Improved |
| add continuous remeshing after DMTet (128) (ablation) | DMTet only · DTU | Chamfer distance: 6.94 -> 1.57 | Improved |
| increase DMTet resolution 128 to 256 (ablation) | DMTet only · DTU | Chamfer distance: 6.94 -> 4.20 | Improved |
| use dense mesh instead of default density (ablation) | Full model · DTU | Chamfer distance: 1.57 -> 1.67 | No clear effect |
| use sparse mesh instead of default density (ablation) | Full model · DTU | Chamfer distance: 1.57 -> 1.66 | No clear effect |
| compare to IMLS-Splatting | Mesh Splatting · DTU | Chamfer distance (cm): 0.57 -> 0.62 | Hurt |
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.
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · better on 2 of 2 shared results
- NeuS2 — NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction (ICCV 2023) · 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
- Volumetric Surfaces: Representing Fuzzy Geometries with Layered Meshes (CVPR 2025 · similarity 0.90)
- Tetrahedron Splatting for 3D Generation (NeurIPS 2024 · similarity 0.89)
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (CVPR 2024 · similarity 0.88)
- Geometry Field Splatting with Gaussian Surfels (CVPR 2025 · similarity 0.88)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.88)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.87)
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.
