Research map / Point clouds / Shape and surface completion
Deep Marching Cubes: Learning Explicit Surface Representations
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 |
|---|---|---|---|
| Predict an explicit mesh with differentiable surface-based supervision instead of an implicit representation followed by marching cubes | Deep Marching Cubes · ShapeNet 3D shape prediction | Chamfer: 0.354 -> 0.218 | Helped |
| Add the occupancy loss to the point-to-mesh loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Hamming: 83.69% -> 4.67% | Helped |
| Add the occupancy loss to the point-to-mesh loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Chamfer: 0.339 -> 0.357 | Hurt |
| Add the smoothness loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Chamfer: 0.357 -> 0.240 | Helped |
| Add the smoothness loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Accuracy: 0.429 -> 0.224 | Helped |
| Add the smoothness loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Hamming: 4.67% -> 0.56% | Helped |
| Add the curvature loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Chamfer: 0.240 -> 0.245 | Hurt |
| Add the curvature loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Accuracy: 0.224 -> 0.219 | Helped |
| Add the curvature loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Completeness: 0.255 -> 0.272 | Hurt |
| Add the curvature loss (ablation) | Deep Marching Cubes, 2D car silhouettes · ShapeNet 2D car silhouettes | Hamming: 0.56% -> 0.53% | Helped |
| Add noise with σ = 0.15 to the ground-truth points | Deep Marching Cubes · ShapeNet 2D car silhouettes | Chamfer: 0.245 -> 0.246 | No clear effect |
| Add noise with σ = 0.30 to the ground-truth points | Deep Marching Cubes · ShapeNet 2D car silhouettes | Chamfer: 0.245 -> 0.296 | Hurt |
Closest papers by meaning
- Analytic Marching: An Analytic Meshing Solution from Deep Implicit Surface Networks (ICML 2020 · similarity 0.89)
- Learning to Infer Implicit Surfaces without 3D Supervision (NeurIPS 2019 · similarity 0.89)
- SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (CVPR 2017 · similarity 0.89)
- DeepCurrents: Learning Implicit Representations of Shapes With Boundaries (CVPR 2022 · similarity 0.88)
- Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis (CVPR 2017 · similarity 0.88)
- NeAT: Learning Neural Implicit Surfaces With Arbitrary Topologies From Multi-View Images (CVPR 2023 · 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.