Research map / Point clouds / Point cloud classification
RigidFlow: Self-Supervised Scene Flow Learning on Point Clouds by Local Rigidity Prior
Check 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 |
|---|---|---|---|
| Generate pseudo labels using locally rigid region alignments rather than nearest-neighbor matches (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.217 -> 0.071 | Helped |
| Use region-wise rigid alignment rather than region-wise center alignment (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.089 -> 0.071 | Helped |
| Initialize point mapping with predicted flow (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.356 -> 0.071 | Helped |
| Generate pseudo flow from rigid motion rather than correspondence coordinate differences (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.080 -> 0.071 | Helped |
| Increase update iterations from 1 to 4 (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.077 -> 0.069 | Helped |
| Vary the number of supervoxels (ablation) | RigidFlow · FT3Ds testing set | EPE: 0.074, 0.073, 0.071, 0.077 | Mixed |
| Train self-supervised with RigidFlow rather than the strongest reported self-supervised baseline | FLOT on FT3Ds testing set · FT3Ds testing set | EPE: 0.0852 -> 0.0692 | Helped |
| Train self-supervised with RigidFlow rather than the strongest reported self-supervised baseline | FLOT on KITTIs · KITTIs | EPE: 0.1021 -> 0.0619 | Helped |
| Train on unannotated KITTI data with RigidFlow rather than supervised FLOT trained on synthetic data | FLOT on KITTIo · KITTIo | EPE: 0.107 -> 0.102 | Helped |
| Train self-supervised on KITTIr with RigidFlow rather than WWL | FLOT on KITTIt · KITTIt | EPE: 0.169 -> 0.117 | Helped |
Closest papers by meaning
- Self-Point-Flow: Self-Supervised Scene Flow Estimation From Point Clouds With Optimal Transport and Random Walk (CVPR 2021 · similarity 0.92)
- Self-Supervised Robust Scene Flow Estimation via the Alignment of Probability Density Functions (AAAI 2022 · similarity 0.91)
- VoteFlow: Enforcing Local Rigidity in Self-Supervised Scene Flow (CVPR 2025 · similarity 0.90)
- 3DSFLabelling: Boosting 3D Scene Flow Estimation by Pseudo Auto-labelling (CVPR 2024 · similarity 0.90)
- ZeroFlow: Scalable Scene Flow via Distillation (ICLR 2024 · similarity 0.89)
- Rectified Point Flow: Generic Point Cloud Pose Estimation (NeurIPS 2025 · similarity 0.87)
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.