Research map / 3D vision / Gaussian splatting
MeshSplatting: Differentiable Rendering with Opaque Meshes
Jan Held, Sanghyun Son, Renaud Vandeghen, Daniel Rebain, Matheus Gadelha, Yi Zhou, Anthony Cioppa, Ming C. Lin, Marc Van Droogenbroeck, Andrea Tagliasacchi
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 |
|---|---|---|---|
| replace spherical harmonics colors with plain RGB vertex colors (ablation) | MeshSplatting · Mip-NeRF360 | PSNR: -2.07 dB | Hurt |
| add normal loss Ln (ablation) | MeshSplatting · Mip-NeRF360 | PSNR: -0.10 dB (removal gives +0.10) | Hurt |
| add depth loss Ld (ablation) | MeshSplatting · Mip-NeRF360 | PSNR: -0.05 dB (removal gives +0.05) | No clear effect |
| add depth alignment loss Lz (ablation) | MeshSplatting · Mip-NeRF360 | PSNR: -0.02 dB (removal gives +0.02) | No clear effect |
| prune triangles immediately after restricted Delaunay instead of after training (ablation) | MeshSplatting · Garden scene connectivity | mean neighbors: 3.7 -> 2.5 | Hurt |
| use MeshSplatting instead of MiLo | mesh-based NVS · Mip-NeRF360 | training time (min): 106m -> 48m | Improved |
| use MeshSplatting instead of MiLo | mesh-based NVS · Tanks & Temples | PSNR: 21.46 -> 20.52 | Hurt |
| use MeshSplatting instead of MiLo | mesh-based NVS · Mip-NeRF360 | PSNR: 24.09 -> 24.78 | Improved |
| use opaque triangles only (Triangle Splatting opaque vs MeshSplatting connected mesh) | triangle-based representation · Mip-NeRF360 | PSNR: 21.05 -> 24.78 | Improved |
Later papers that compared with this method
1 later accepted paper reports MeshSplatting as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- T$^2$-Splat: Adaptive Topology Mesh Splatting with Texture Residuals (NeurIPS 2026 · better on 7 of 7 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
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (CVPR 2024 · similarity 0.90)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.90)
- Mani-GS: Gaussian Splatting Manipulation with Triangular Mesh (CVPR 2025 · similarity 0.90)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.90)
- H3D-DGS: Exploring Heterogeneous 3D Motion Representation for Deformable 3D Gaussian Splatting (NeurIPS 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.
