Research map / 3D vision / Face and shape
SE(n)-Invariant Flow Matching: A General Framework with Application to Object Reassembly
Gaël Heck, Sylvie Le Hégarat-Mascle, Nicolas Lermé
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 canonical gauge fixing (ablation) | SE(2) flow matching · SVJP | Overlap: .80 -> .90 | Improved |
| add quotient-invariant loss (ablation) | SE(2) flow matching · SVJP | Overlap: .80 -> .92 | Improved |
| add quotient loss on top of gauge fixing (ablation) | SE(2) flow matching · SVJP | Overlap: .90 -> .93 | Improved |
| replace gauge fixing with multi-anchor supervision (ablation) | SE(2) flow matching · SVJP | Overlap: .90 -> .87 | Hurt |
| add multi-anchor supervision to base (ablation) | SE(2) flow matching · SVJP | Overlap: .80 -> .87 | Improved |
| add test-time prediction projection (ablation) | SE(2) flow matching · SVJP | Overlap: .93 -> .94 | Improved |
| add gauge fixing (ablation) | SE(3) flow matching · Breaking Bad Everyday | RMSE(R): 9.67 -> 6.98 | Improved |
| add quotient loss (ablation) | SE(3) flow matching · Breaking Bad Everyday | RMSE(R): 9.67 -> 6.07 | Improved |
| add gauge fixing and quotient loss (ablation) | SE(3) flow matching · Breaking Bad Artifact | RMSE(R): 10.02 -> 5.41 | Improved |
| use invariant flow matching instead of strongest SE(2) baseline at 1% noise | SVFP · SVFP | Overlap: 0.76 to 0.83 | Improved |
| use invariant flow matching instead of GARF | SE(3) reassembly · Breaking Bad Everyday | RMSE(R): 6.43 to 5.80 | Improved |
| use invariant flow matching instead of Islam et al. | RePAIR · RePAIR | Overlap: 0.21 -> 0.41 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GARF — GARF: Learning Generalizable 3D Reassembly for Real-World Fractures (ICCV 2025) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SIGMA: Scale-Invariant Global Sparse Shape Matching (ICCV 2023 · similarity 0.88)
- Chebyshev Differential Flows for Shape Matching and Interpolation with Endpoint Guidance (NeurIPS 2026 · similarity 0.87)
- Fast Globally Optimal and Geometrically Consistent 3D Shape Matching (ICCV 2025 · similarity 0.87)
- Shape Non-rigid Kinematics (SNK): A Zero-Shot Method for Non-Rigid Shape Matching via Unsupervised Functional Map Regularized Reconstruction (NeurIPS 2023 · similarity 0.87)
- Rectified Point Flow: Generic Point Cloud Pose Estimation (NeurIPS 2025 · similarity 0.85)
- Flow Matching on General Geometries (ICLR 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.
