Research map / Point clouds / Point cloud classification
CRIN: Rotation-Invariant Point Cloud Analysis and Rotation Estimation via Centrifugal Reference Frame
Read the paperCheck 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 |
|---|---|---|---|
| Replace ordinary point coordinates with rotation-invariant CRF features (ablation) | CRIN · ModelNet40 z/SO(3) | accuracy (%): 16.8 -> 89.1 | Helped |
| Use a complete CRF rather than PCRF alone (ablation) | CRIN · ModelNet40 z/SO(3) | accuracy (%): 77.0 -> 89.1 | Helped |
| Add continuous distribution of rotations (ablation) | CRIN with CRF · ModelNet40 z/SO(3) | accuracy (%): 89.1 -> 90.9 | Helped |
| Add attention-based sampling (ablation) | CRIN with CRF · ModelNet40 z/SO(3) | accuracy (%): 89.1 -> 89.3 | Helped |
| Add relation module (ablation) | CRIN with CRF · ModelNet40 z/SO(3) | accuracy (%): 89.1 -> 90.0 | Helped |
| Use all components together (ablation) | CRIN · ModelNet40 z/SO(3) | accuracy (%): 90.9 -> 91.8 | Helped |
| Use CRIN rather than the strongest listed baseline | CRIN · ModelNet40 z/SO(3) | accuracy (%): 91.6 -> 91.8 | Helped |
| Use CRIN rather than the strongest listed baseline | CRIN · ShapeNet part z/SO(3) | mean IoU (%): 79.4 -> 80.5 | Helped |
| Use CRIN rather than the strongest listed baseline | CRIN · ScanObjectNN z*/SO(3) | accuracy (%): 86.7 -> 84.7 | Hurt |
| Estimate rotation from an anchor selected by CRIN without rotation supervision | CRIN · ModelNet40 rotation estimation | Accuracy (%): 98.1 -> 98.9 | Helped |
| Use CRIN rather than SpinNet for rotation estimation | CRIN · ModelNet40 rotation estimation | Mean AD: 0.0672 -> 0.0678 | Hurt |
Closest papers by meaning
- Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution (AAAI 2020 · similarity 0.89)
- PaRot: Patch-Wise Rotation-Invariant Network via Feature Disentanglement and Pose Restoration (AAAI 2023 · similarity 0.88)
- ClusterNet: Deep Hierarchical Cluster Network With Rigorously Rotation-Invariant Representation for Point Cloud Analysis (CVPR 2019 · similarity 0.88)
- Local-consistent Transformation Learning for Rotation-invariant Point Cloud Analysis (CVPR 2024 · similarity 0.86)
- Leveraging 3D Geometric Priors in 2D Rotation Symmetry Detection (CVPR 2025 · similarity 0.86)
- Learning with 3D rotations, a hitchhiker's guide to SO(3) (ICML 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.