Research map / 3D vision / Face and shape
HiPPI: Higher-Order Projected Power Iterations for Scalable Multi-Matching
Florian Bernard, Johan Thunberg, Paul Swoboda, Christian Theobalt
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 QuickMatch initialisation instead of random initialisation (ablation) | HiPPI · HiPPI dataset | fscore: 0.90 -> 0.92 | Improved |
| replace QuickMatch with HiPPI (adds geometric consistency) | multi-image matching · HiPPI dataset | fscore: 0.71 -> 0.92 | Improved |
| random-init HiPPI vs QuickMatch (ablation) | HiPPI · HiPPI dataset | fscore: 0.71 -> 0.90 | Improved |
| add geometric relations via higher-order term (HiPPI vs Spectral synchronisation) | permutation synchronisation · Willow (Motorbike) | fscore: 0.37 -> 0.84 | Improved |
| HiPPI vs Spectral on Face | permutation synchronisation · Willow (Face) | fscore: 0.98 -> 1.00 | Improved |
| HiPPI vs Wang et al. (geometry-aware baseline) | Wang et al. · Willow (Motorbike) | fscore: 0.65 -> 0.84 | Improved |
| HiPPI vs Wang et al. on Car | Wang et al. · Willow (Car) | fscore: 0.71 -> 0.74 | Improved |
| HiPPI vs Yan et al. runtime (Car) | Yan et al. · Willow (Car) | runtime: 21.5s -> 0.6s | Improved |
| QuickMatch instead of pairwise on bijective matching | Pairwise · Willow (Car) | fscore: 0.54 -> 0.24 | Hurt |
Closest papers by meaning
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · similarity 0.89)
- Fast Globally Optimal and Geometrically Consistent 3D Shape Matching (ICCV 2025 · similarity 0.88)
- DiscoMatch: Fast Discrete Optimisation for Geometrically Consistent 3D Shape Matching (ECCV 2024 · similarity 0.88)
- Higher-Order Ratio Cycles for Fast and Globally Optimal Shape Matching (CVPR 2025 · similarity 0.87)
- Partial-to-Partial Shape Matching with Geometric Consistency (CVPR 2024 · similarity 0.86)
- Towards Optimizing Large-Scale Multi-Graph Matching in Bioimaging (CVPR 2025 · 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.
