Research map / 3D vision / Face and shape
Exact Shape Correspondence via 2D graph convolution
Barakeel Fanseu Kamhoua, Lin Zhang, Yongqiang Chen, Han Yang, MA KAILI, Bo Han, Bo Li, James Cheng
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 |
|---|---|---|---|
| use 2D graph convolution (2D-GEM) instead of GRAMPA | exact shape matching · TOPKIDS geodesic error 0 | accuracy: 84.6% -> 90.5% | Improved |
| use 2-hop instead of 1-hop adjacency refinement in GRAMPA | GRAMPA · TOPKIDS geodesic error 0 | accuracy: 60% -> 84.6% | Improved |
| 2D-GEM with LMD and 500 eigenvectors instead of DIR-1000 | nearly-isometric matching · TOSCA geodesic error 0 | accuracy: 59.8% -> 92.5% | Improved |
| 2D-GEM with LMD and 500 eigenvectors instead of DIR-1000 | nearly-isometric matching · SCAPE geodesic error 0 | accuracy: 69.5% -> 84.9% | Improved |
| apply GRAMPA to nearly-isometric shapes | GRAMPA · SCAPE geodesic error 0 | accuracy: 84.6% -> 10% | Hurt |
| use more eigenvectors or more iterations | 2D-GEM · shape matching | accuracy and running time | Mixed |
| compute only a subset of eigenvectors instead of full spectrum | 2D-GEM vs GRAMPA · TOPKIDS and SCAPE | running time | Improved |
| set LMD thresholds arbitrarily large (100) for non-isometric shapes | 2D-GEM · TOPKIDS | accuracy | Improved |
Closest papers by meaning
- Efficient and Robust Shape Correspondence via Sparsity-Enforced Quadratic Assignment (CVPR 2020 · similarity 0.88)
- 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)
- HOPE: Shape Matching Via Aligning Different K-hop Neighbourhoods (NeurIPS 2024 · similarity 0.85)
- Stable-SCore: A Stable Registration-based Framework for 3D Shape Correspondence (CVPR 2025 · similarity 0.85)
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · 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.
