Research map / 3D vision / Face and shape
VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids
Albert Pumarola, Artsiom Sanakoyeu, Lior Yariv, Ali Thabet, Yaron Lipman
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 normal loss (vs. w/o normals) (ablation) | VisCo Grids · Benchmark Anchor | d_C: 0.61 -> 0.21 | Improved |
| add viscosity loss (vs. w/o viscosity) (ablation) | VisCo Grids · Benchmark Anchor | d_H: 10.83 -> 3.00 | Improved |
| add coarea loss (vs. w/o coarea) (ablation) | VisCo Grids · Benchmark Anchor | d_C: 0.72 -> 0.21 | Improved |
| add coarea loss (vs. w/o coarea) (ablation) | VisCo Grids · Benchmark DC | d_H: 6.58 -> 2.22 | Improved |
| add coarea loss (vs. w/o coarea) (ablation) | VisCo Grids · Benchmark Daratech | d_H: 2.19 -> 4.06 | Hurt |
| add viscosity loss (vs. w/o viscosity) (ablation) | VisCo Grids · Benchmark Gargoyle | d_C: 0.47 -> 0.17 | Improved |
| add normal loss (vs. w/o normals) (ablation) | VisCo Grids · Benchmark Gargoyle | d_H: 6.32 -> 4.40 | Improved |
| add normal loss (vs. w/o normals) (ablation) | VisCo Grids · Benchmark Lord Quas | d_C: 0.12 -> 0.12 | No clear effect |
| add viscosity loss (vs. w/o viscosity) (ablation) | VisCo Grids · Benchmark DC | d_C: 0.15 -> 0.15 | No clear effect |
| reduce training time budget from 30 min to 8 min (ablation) | VisCo Grids · Benchmark mean | d_C: 0.19 -> 0.20 | No clear effect |
| reduce training time budget from 30 min to 8 min (ablation) | VisCo Grids · Benchmark mean | d_H: 2.95 -> 3.53 | Hurt |
| reduce training time budget from 30 min to 8 min (ablation) | VisCo Grids · Benchmark Anchor | d_H: 3.00 -> 5.69 | Hurt |
Closest papers by meaning
- GridPull: Towards Scalability in Learning Implicit Representations from 3D Point Clouds (ICCV 2023 · similarity 0.88)
- Phase Transitions, Distance Functions, and Implicit Neural Representations (ICML 2021 · similarity 0.88)
- GridFormer: Point-Grid Transformer for Surface Reconstruction (AAAI 2024 · similarity 0.87)
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals (NeurIPS 2025 · similarity 0.87)
- NeuralIndicator: Implicit Surface Reconstruction from Neural Indicator Priors (ICML 2024 · similarity 0.86)
- NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views (AAAI 2024 · similarity 0.86)
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.
