Research map / 3D vision / Face and shape
Analytic Marching: An Analytic Meshing Solution from Deep Implicit Surface Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 discretized marching cubes with analytic marching | D6-W60 SDF mesh extraction · Five ShapeNetCoreV1 categories, mean | CD (×10^-1): 5.5730 -> 5.5049 | Helped |
| Replace discretized marching cubes with analytic marching | D6-W60 SDF mesh extraction · Five ShapeNetCoreV1 categories, mean | Time (sec.): 2.41 -> 20.8 | Hurt |
| Increase SDF network capacity from 360 to 900 neurons (ablation) | Analytic Marching on Rifle · Rifle | CD (×10^-1): 5.29 -> 3.64 | Helped |
| Increase SDF network capacity from 360 to 900 neurons (ablation) | Analytic Marching on Rifle · Rifle | Time (sec.): 15.0 -> 173 | Hurt |
| At 360 neurons, use D6-W60 instead of D4-W90 (ablation) | Analytic Marching on Rifle · Rifle | CD (×10^-1): 6.16 -> 5.29 | Helped |
| At 360 neurons, use D6-W60 instead of D8-W45 (ablation) | Analytic Marching on Rifle · Rifle | CD (×10^-1): 5.67 -> 5.29 | Helped |
| At 360 neurons, use D8-W45 instead of D6-W60 (ablation) | Analytic Marching on Rifle · Rifle | Time (sec.): 15.0 -> 9.75 | Helped |
| At 900 neurons, use D10-W90 instead of D15-W60 (ablation) | Analytic Marching on Rifle · Rifle | CD (×10^-1): 3.85 -> 3.64 | Helped |
| At 900 neurons, use D10-W90 instead of D20-W45 (ablation) | Analytic Marching on Rifle · Rifle | CD (×10^-1): 4.21 -> 3.64 | Helped |
| At 900 neurons, use D20-W45 instead of D10-W90 (ablation) | Analytic Marching on Rifle · Rifle | Time (sec.): 173 -> 47.3 | Helped |
Closest papers by meaning
- Deep Marching Cubes: Learning Explicit Surface Representations (CVPR 2018 · similarity 0.89)
- Iso-Points: Optimizing Neural Implicit Surfaces With Hybrid Representations (CVPR 2021 · similarity 0.88)
- NeAT: Learning Neural Implicit Surfaces With Arbitrary Topologies From Multi-View Images (CVPR 2023 · similarity 0.87)
- Self-Supervised Dual Contouring (CVPR 2024 · similarity 0.86)
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (CVPR 2025 · similarity 0.85)
- Polyhedral Complex Derivation from Piecewise Trilinear Networks (NeurIPS 2024 · similarity 0.84)
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.