Research map / 3D vision / Face and shape
DiscoMatch: Fast Discrete Optimisation for Geometrically Consistent 3D Shape Matching
Paul Roetzer, Ahmed Abbas, Dongliang Cao, Florian Bernard, Paul Swoboda
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 |
|---|---|---|---|
| replace elastic energy with the proposed energy in the ILP (ablation) | SMCOMB · FAUST (450 faces) | mean geodesic error: 0.087 -> 0.042 | Improved |
| replace elastic energy with the proposed energy in the ILP (ablation) | FastDOG · FAUST (450 faces) | runtime (s): 925 -> 327 | Improved |
| replace elastic energy with the proposed energy in the ILP (ablation) | DiscoMatch · FAUST (450 faces) | runtime (s): 423 -> 134 | Improved |
| replace elastic energy with the proposed energy in the ILP (ablation) | DiscoMatch · FAUST (450 faces) | number of infeasible solutions: 8 -> 2 | Improved |
| add quasi-Newton updates and BDD constraint splitting to dual solver (ablation) | DiscoMatch · Convergence over 55 instances (500 to 1000 triangles) | runtime of dual optimisation: up to 11x faster | Improved |
| add geometric consistency constraints | DiscoMatch · FAUST, DT4D, SHREC'20 | mean geodesic error: sometimes over 50% decrease | Improved |
| add geometric consistency constraints | DiscoMatch · SMAL | mean geodesic error | Hurt |
| use permutation-matrix search (Kernel matching) for different discretisations | Kernel matching baseline · all datasets | matching quality | Hurt |
| use coarse-to-fine pruning to scale to 4k triangles | DiscoMatch · FAUST, DT4D-Intra | runtime: about 20 minutes vs few seconds for Cao et al. | Mixed |
| apply solver improvements to other ILPs beyond shape matching | DiscoMatch solver · difficult ILP datasets | runtime: 5-10 times improvement | Improved |
Closest papers by meaning
- Partial-to-Partial Shape Matching with Geometric Consistency (CVPR 2024 · similarity 0.91)
- SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency (CVPR 2024 · similarity 0.91)
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · similarity 0.88)
- 3D Geometric Shape Assembly via Efficient Point Cloud Matching (ICML 2024 · similarity 0.88)
- Higher-Order Ratio Cycles for Fast and Globally Optimal Shape Matching (CVPR 2025 · similarity 0.87)
- 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.
