Research map / Point clouds / Registration and matching
Rethinking Rotation Invariance with Point Cloud Registration
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 |
|---|---|---|---|
| Add self-attention angular embedding (ablation) | Aligned Integration Transformer · ModelNet40 z/SO(3) | accuracy: 90.0 -> 90.6 | Helped |
| Add cross-attention angular embedding (ablation) | Aligned Integration Transformer · ModelNet40 z/SO(3) | accuracy: 90.0 -> 90.2 | Helped |
| Use both self- and cross-attention angular embeddings (ablation) | Aligned Integration Transformer · ModelNet40 z/SO(3) | accuracy: 90.0 -> 91.0 | Helped |
| Add intra-frame aligned self-attention to global features (ablation) | Aligned Integration Transformer · ModelNet40 z/SO(3) | accuracy: 91.0 -> 90.2 | Hurt |
| Replace combined attention-logit integration with sequential repeated self- and cross-attention (ablation) | Aligned Integration Transformer · ModelNet40 z/SO(3) | accuracy: 91.0 -> 90.4 | Hurt |
| Add both local and global registration losses (ablation) | Deep feature registration · ModelNet40 z/SO(3) | accuracy: 90.0 -> 91.0 | Helped |
| Add global registration loss (ablation) | Deep feature registration · ModelNet40 z/SO(3) | accuracy: 90.0 -> 90.6 | Helped |
| Add local registration loss (ablation) | Deep feature registration · ModelNet40 z/SO(3) | accuracy: 90.0 -> 90.2 | Helped |
Closest papers by meaning
- Rotation-Invariant Transformer for Point Cloud Matching (CVPR 2023 · similarity 0.91)
- Local-consistent Transformation Learning for Rotation-invariant Point Cloud Analysis (CVPR 2024 · similarity 0.89)
- Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detection (NeurIPS 2025 · similarity 0.88)
- RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning (AAAI 2025 · similarity 0.88)
- Invaria: Learning Scale and Density Invariance in Point Clouds via Next-Resolution Prediction (NeurIPS 2026 · similarity 0.87)
- Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform (CVPR 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.