Research map / Graph learning / Graph algorithms and clustering
G-MSM: Unsupervised Multi-Shape Matching With Graph-Based Affinity Priors
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 |
|---|---|---|---|
| Add graph-based multi-matching module III (ablation) | G-MSM · FAUST | mean geodesic error: 1.7 -> 1.5 | Helped |
| Add graph-based multi-matching module III (ablation) | G-MSM · SCAPE | mean geodesic error: 3.3 -> 1.8 | Helped |
| Add graph-based multi-matching module III (ablation) | G-MSM · SURREAL | mean geodesic error: 8.1 -> 2.1 | Helped |
| Add graph-based multi-matching module III (ablation) | G-MSM · SHREC'19 Connectivity | mean geodesic error: 6.2 -> 2.7 | Helped |
| Use an MST graph instead of the full graph (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 8.62 | Hurt |
| Use a star graph instead of the full graph (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 11.92 | Hurt |
| Replace DiffusionNet features with SHOT features (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 7.93 | No clear effect |
| Replace DiffusionNet features with GraphNN features (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 14.18 | Hurt |
| Replace DeepShells matching with a functional map layer (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 39.12 | Hurt |
| Replace DeepShells matching with a single Sinkhorn layer (ablation) | G-MSM · TOPKIDS | mean geodesic error: 7.92 -> 12.25 | Hurt |
Closest papers by meaning
- DcMatch: Unsupervised Multi-Shape Matching with Dual-Level Consistency (AAAI 2026 · similarity 0.91)
- Probabilistic Deformation Consistency for Unsupervised Shape Matching (AAAI 2026 · similarity 0.90)
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · similarity 0.87)
- Partial-to-Partial Shape Matching with Geometric Consistency (CVPR 2024 · similarity 0.87)
- GenCorres: Consistent Shape Matching via Coupled Implicit-Explicit Shape Generative Models (ICLR 2024 · similarity 0.86)
- HOPE: Shape Matching Via Aligning Different K-hop Neighbourhoods (NeurIPS 2024 · similarity 0.86)
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.