Research map / Point clouds / Point cloud classification
Self-Supervised 3D Scene Flow Estimation and Motion Prediction Using Local Rigidity Prior
Ruibo Li, Chi Zhang, Zhe Wang, Chunhua Shen, Guosheng Lin
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 region-wise rigid alignment instead of nearest-neighbor matching (ablation) | RigidFlow · FT3Ds test | EPE: 0.242 -> 0.059 | Improved |
| add region-wise rigid alignment instead of nearest-neighbor matching (ablation) | RigidFlow · FT3Do test | EPE: 0.734 -> 0.210 | Improved |
| use full rigid alignment instead of only center alignment (ablation) | RigidFlow · FT3Ds test | EPE: 0.100 -> 0.059 | Improved |
| use full rigid alignment instead of only center alignment (ablation) | RigidFlow · FT3Do test | EPE: 0.225 -> 0.210 | Improved |
| add confidence reweighting from forward-backward consistency (ablation) | RigidFlow · FT3Do test | EPE: 0.210 -> 0.173 | Improved |
| add validity reasoning for pseudo labels (ablation) | RigidFlow · FT3Ds test | EPE: 0.059 -> 0.051 | Improved |
| add validity reasoning for pseudo labels (ablation) | RigidFlow · FT3Do test | EPE: 0.210 -> 0.167 | Improved |
| add both confidence reweighting and validity reasoning (RigidFlow++) (ablation) | RigidFlow · FT3Ds test | EPE: 0.059 -> 0.047 | Improved |
| add both confidence reweighting and validity reasoning (RigidFlow++) (ablation) | RigidFlow · FT3Do test | EPE: 0.210 -> 0.144 | Improved |
| increase update iterations from 1 to 4 (ablation) | RigidFlow++ · FT3Ds test | EPE: 0.052 -> 0.047 | Improved |
| increase update iterations from 1 to 4 (ablation) | RigidFlow++ · FT3Do test | EPE: 0.152 -> 0.144 | Improved |
| increase supervoxel number from 10 to 30 (ablation) | RigidFlow++ · FT3Ds test | EPE: 0.057 -> 0.047 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SCOOP — SCOOP: Self-Supervised Correspondence and Optimization-Based Scene Flow (CVPR 2023) · better on 1 of 2 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
- RigidFlow: Self-Supervised Scene Flow Learning on Point Clouds by Local Rigidity Prior (CVPR 2022 · similarity 0.97)
- Self-Point-Flow: Self-Supervised Scene Flow Estimation From Point Clouds With Optimal Transport and Random Walk (CVPR 2021 · similarity 0.91)
- 3DSFLabelling: Boosting 3D Scene Flow Estimation by Pseudo Auto-labelling (CVPR 2024 · similarity 0.91)
- VoteFlow: Enforcing Local Rigidity in Self-Supervised Scene Flow (CVPR 2025 · similarity 0.89)
- ZeroFlow: Scalable Scene Flow via Distillation (ICLR 2024 · similarity 0.88)
- Self-Supervised Class-Agnostic Motion Prediction with Spatial and Temporal Consistency Regularizations (CVPR 2024 · 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.
