Research map / 3D vision / Face and shape
Neural Convolutional Surfaces
Check 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 |
|---|---|---|---|
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Armadillo | Chamfer distance (×10^3): 1.06 -> 0.54 | Helped |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Bimba | Chamfer distance (×10^3): 2.07 -> 1.04 | Helped |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Dino | Chamfer distance (×10^3): 1.55 -> 1.48 | Helped |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Dragon | Chamfer distance (×10^3): 0.62 -> 0.57 | Helped |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Grog | Chamfer distance (×10^3): 0.81 -> 1.28 | Hurt |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Seahorse | Chamfer distance (×10^3): 1.15 -> 0.44 | Helped |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Elephant | Chamfer distance (×10^3): 2.24 -> 2.49 | Hurt |
| Use NCS instead of the strongest reported baseline | 100K-parameter shape representations · Gargoyl | Chamfer distance (×10^3): 6.30 -> 2.29 | Helped |
| Add vector displacements rather than restricting offsets to surface normals (ablation) | NCS · Shape reconstruction | Reconstruction quality | Helped |
| Add the learned coarse model rather than relying on PCA frames alone (ablation) | NCS · Shape reconstruction | Reconstruction quality | Helped |
| Use CNN upsampling (ablation) | NCS fine model · Surface detail reconstruction | Detail reconstruction | Helped |
Closest papers by meaning
- CoFie: Learning Compact Neural Surface Representations with Coordinate Fields (NeurIPS 2024 · similarity 0.90)
- Neural Surface Maps (CVPR 2021 · similarity 0.90)
- A Papier-Mâché Approach to Learning 3D Surface Generation (CVPR 2018 · similarity 0.89)
- SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (CVPR 2017 · similarity 0.89)
- Improving Neural Network Surface Processing with Principal Curvatures (NeurIPS 2024 · similarity 0.89)
- Flatten Anything: Unsupervised Neural Surface Parameterization (NeurIPS 2024 · 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.