Research map / 3D vision / Human pose estimation
Leveraging SE(3) Equivariance for Self-supervised Category-Level Object Pose Estimation from Point Clouds
Xiaolong Li, Yijia Weng, Li Yi, Leonidas Guibas, A. Lynn Abbott, Shuran Song, He Wang
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 |
|---|---|---|---|
| use non-equivariant KPConv instead of equivariant EPN for canonical shape reconstruction (ablation) | self-supervised pose estimation · ModelNet40 partial airplane | 5deg accuracy: 0.95 -> 0.05 | Hurt |
| use equivariant EPN instead of KPConv for direct pose regression (ablation) | self-supervised pose estimation · ModelNet40 partial airplane | 5deg accuracy: 0.93 -> 0.95 | Improved |
| replace SE(3) Transformer with EPN-60 backbone (ablation) | self-supervised pose estimation · ModelNet40 partial airplane | 5deg accuracy: 0.46 -> 0.95 | Improved |
| use EPN-60 instead of EPN-20 sub-region division (ablation) | self-supervised pose estimation · ModelNet40 partial airplane | 5deg accuracy: 0.50 -> 0.95 | Improved |
| EPN-60 instead of SE(3) Transformer on complete shapes (ablation) | self-supervised pose estimation · ModelNet40 complete airplane | mean rotation error (deg): 6.65 -> 23.09 | Mixed |
| self-supervised method versus supervised EPN on complete chair | self-supervised pose estimation · ModelNet40 complete chair | mean rotation error (deg): 8.56 -> 7.05 | Improved |
| self-supervised method versus ICP on partial car | self-supervised pose estimation · ModelNet40 partial car | 5deg rotation accuracy: 0.36 -> 0.12 | Hurt |
| self-supervised fine-tuning on real NOCS data (ablation) | pose estimation · NOCS-REAL275 | 5deg accuracy | Improved |
| self-supervised method versus ICP on NOCS-REAL275 | pose estimation · NOCS-REAL275 | 5deg accuracy | Improved |
Closest papers by meaning
- Self-Supervised Category-Level Articulated Object Pose Estimation with Part-Level SE(3) Equivariance (ICLR 2023 · similarity 0.93)
- SE(3)-Equivariance with Geometric and Topological Guidance for Category-Level Object Pose Estimation (CVPR 2026 · similarity 0.91)
- E2PN: Efficient SE(3)-Equivariant Point Network (CVPR 2023 · similarity 0.90)
- $\mathrm{SE}(3)$-Equivariant Attention Networks for Shape Reconstruction in Function Space (ICLR 2023 · similarity 0.90)
- SIEVE: Overcoming Topological Obstruction in Equivariant Self-Supervised Learning (NeurIPS 2026 · similarity 0.88)
- SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation (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.
