Research map / 3D vision / Face and shape
Coarse-to-fine Contrast: A Hybrid Self-supervised Method for Non-rigid 3D Shape Matching
Feifan Luo, Ting Li, Zhao Li, Hongyang Chen
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 extrinsic (elastic basis) branch losses (ablation) | Ours · TOPKIDS | mean geodesic error (x100): 12.9 -> 4.9 | Improved |
| add extrinsic (elastic basis) branch losses (ablation) | Ours · FAUST->SHREC'19 | mean geodesic error (x100): 4.3 -> 4.4 | No clear effect |
| add extrinsic orthogonality and bijectivity losses L_fmap (ablation) | Ours · FAUST->SHREC'19 | mean geodesic error (x100): 5.3 -> 4.4 | Improved |
| add extrinsic orthogonality and bijectivity losses L_fmap (ablation) | Ours · SMAL | mean geodesic error (x100): 3.6 -> 2.9 | Improved |
| add extrinsic contrastive energies (ablation) | Ours · TOPKIDS | mean geodesic error (x100): 11.3 -> 4.9 | Improved |
| add extrinsic contrastive energies (ablation) | Ours · FAUST->SHREC'19 | mean geodesic error (x100): 4.8 -> 4.4 | Improved |
| add intrinsic contrastive energies (ablation) | Ours · TOPKIDS | mean geodesic error (x100): 26.8 -> 12.9 | Improved |
| add intrinsic contrastive energies (ablation) | Ours · SMAL | mean geodesic error (x100): 5.3 -> 3.3 | Improved |
| add fine properness (spatial-spectral contrastive) loss | ULRSSM · FAUST->SHREC'19 | mean geodesic error (x100): 14.5 -> 7.1 | Improved |
| add fine properness (spatial-spectral contrastive) loss | ULRSSM · TOPKIDS | mean geodesic error (x100): 9.4 -> 6.5 | Improved |
| add fine properness (spatial-spectral contrastive) loss | ULRSSM · SMAL | mean geodesic error (x100): 4.5 -> 4.6 | No clear effect |
| add spectral contrastive loss | SURFMNet · TOPKIDS | mean geodesic error (x100): 40.7 -> 21.1 | Improved |
Closest papers by meaning
- Hybrid Functional Maps for Crease-Aware Non-Isometric Shape Matching (CVPR 2024 · similarity 0.87)
- Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation (CVPR 2024 · similarity 0.86)
- Partial-to-Partial Shape Matching with Geometric Consistency (CVPR 2024 · similarity 0.86)
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · similarity 0.85)
- Memory-Scalable and Simplified Functional Map Learning (CVPR 2024 · similarity 0.85)
- HOPE: Shape Matching Via Aligning Different K-hop Neighbourhoods (NeurIPS 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.
