Research map / 3D vision / Face and shape
Minimal Neural Atlas: Parameterizing Complex Surfaces with Minimal Charts and Distortion
Weng Fei Low, Gim Hee Lee
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 distortion loss L_dist (ablation) | Minimal Neural Atlas · ShapeNet reconstruction (3 charts) | metric distortion: 10.23 -> 2.189 | Improved |
| add distortion loss L_dist (ablation) | Minimal Neural Atlas · ShapeNet reconstruction (3 charts) | Mesh F@1%: 82.22 -> 82.23 | No clear effect |
| add distortion loss L_dist (ablation) | Minimal Neural Atlas · ShapeNet reconstruction (1 chart) | Point Cloud F@1%: 83.05 -> 80.45 | Hurt |
| add reformulation of occupancy o (ablation) | Minimal Neural Atlas · ShapeNet reconstruction ablation | Mesh F@1%: 67.31 -> 82.23 | Improved |
| add factorization of o~ (PU learning) (ablation) | Minimal Neural Atlas · ShapeNet reconstruction ablation | Mesh F@1%: 27.59 -> 82.23 | Improved |
| add positional encoding on l~ (ablation) | Minimal Neural Atlas · ShapeNet reconstruction ablation | Mesh F@1%: 80.23 -> 82.23 | Improved |
| add positional encoding on l~ (ablation) | Minimal Neural Atlas · ShapeNet reconstruction ablation | metric distortion: 1.848 -> 2.189 | Hurt |
| learnable occupancy domain instead of AtlasNet fixed domain | Minimal Neural Atlas · ShapeNet reconstruction (3 charts) | Mesh CD (1e-4): 10.96 -> 6.761 | Improved |
| learnable domain instead of AtlasNet++ | Minimal Neural Atlas · CLOTH3D++ reconstruction (1 chart) | Mesh CD (1e-4): 4.937 -> 4.476 | No clear effect |
| our distortion-regularized atlas instead of DSP | Minimal Neural Atlas · ShapeNet reconstruction (3 charts) | metric distortion: 0.4130 -> 2.189 | Hurt |
| single-view input with learnable domain | Minimal Neural Atlas · ShapeNet single-view reconstruction (1 chart) | Point Cloud CD (1e-3): 3.100 -> 3.840 | Hurt |
Closest papers by meaning
- Neural Surface Maps (CVPR 2021 · similarity 0.91)
- Neural Convolutional Surfaces (CVPR 2022 · similarity 0.90)
- Coupling Explicit and Implicit Surface Representations for Generative 3D Modeling (ECCV 2020 · similarity 0.88)
- Flatten Anything: Unsupervised Neural Surface Parameterization (NeurIPS 2024 · similarity 0.88)
- Improving Neural Network Surface Processing with Principal Curvatures (NeurIPS 2024 · similarity 0.88)
- CoFie: Learning Compact Neural Surface Representations with Coordinate Fields (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.
