Research map / Point clouds / Shape and surface completion
Point2Skeleton: Learning Skeletal Representations from Point Clouds
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 Point2Skeleton rather than DPC | ShapeNet point-cloud skeletonization · ShapeNet average | CD-Recon: 0.0668 -> 0.0372 | Helped |
| Use Point2Skeleton rather than L1-medial skeleton | ShapeNet point-cloud skeletonization · ShapeNet average | CD-MAT: 0.1665 -> 0.0828 | Helped |
| Add the sampling loss L_s (ablation) | Point2Skeleton skeletal-point prediction · ShapeNet ablation | CD-Recon: 0.3032 -> 0.0525 | Helped |
| Add the point-to-sphere loss L_p (ablation) | Point2Skeleton skeletal-point prediction · ShapeNet ablation | CD-Recon: 0.0535 -> 0.0525 | Helped |
| Add the radius regularizer L_r (ablation) | Point2Skeleton skeletal-point prediction · ShapeNet ablation | CD-Recon: 0.0634 -> 0.0525 | Helped |
| Add graph initialization and GAE (ablation) | Point2Skeleton mesh generation · ShapeNet mesh-generation ablation | CD-Recon: 0.0525 -> 0.0372 | Helped |
| Add the GAE to graph initialization (ablation) | Point2Skeleton mesh generation · ShapeNet mesh-generation ablation | CD-Recon: 0.0383 -> 0.0372 | Helped |
| Add the GAE to graph initialization (ablation) | Point2Skeleton mesh generation · ShapeNet mesh-generation ablation | HD-MAT: 0.1886 -> 0.1898 | Hurt |
| Increase the input sparsity and noise from 2000 points, 0% noise to 500 points, 1% noise (ablation) | Point2Skeleton · ShapeNet input-quality evaluation | CD-Recon: 0.0372 -> 0.0458 | Hurt |
| Increase skeletal point count from N=100 to N=200 (ablation) | Point2Skeleton · ShapeNet skeletal-point-count evaluation | CD-Recon: 0.0372 -> 0.0310 | Helped |
| Increase skeletal point count from N=100 to N=200 (ablation) | Point2Skeleton · ShapeNet skeletal-point-count evaluation | HD-MAT: 0.1898 -> 0.2002 | Hurt |
Closest papers by meaning
- Learnable Skeleton-Aware 3D Point Cloud Sampling (CVPR 2023 · similarity 0.90)
- SPEAL: Skeletal Prior Embedded Attention Learning for Cross-Source Point Cloud Registration (AAAI 2024 · similarity 0.89)
- Skeleton-bridged Point Completion: From Global Inference to Local Adjustment (NeurIPS 2020 · similarity 0.89)
- SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds (ICML 2025 · similarity 0.86)
- Rethinking Point Cloud Data Augmentation: Topologically Consistent Deformation (ICML 2025 · similarity 0.86)
- Unsupervised Template-assisted Point Cloud Shape Correspondence Network (CVPR 2024 · similarity 0.85)
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.