Research map / 3D vision / Human pose estimation
OP-Align: Object-level and Part-level Alignment for Self-supervised Category-level Articulated Object Pose Estimation
Yuchen Che, Ryo Furukawa, Asako Kanezaki
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 object-level alignment before part segmentation, instead of EAP's part-by-part iterative scheme | OP-Align · Synthetic dataset | Segmentation IoU: 68.46 -> 80.70 | Improved |
| replace EAP with OP-Align (no iterative operations) | self-supervised articulated pose estimation · Synthetic dataset | Inference speed (FPS): <1 -> 41 | Improved |
| replace EAP with OP-Align | self-supervised articulated pose estimation · Synthetic dataset | GPU memory (GB): 9.23 -> 2.31 | Improved |
| replace EAP with OP-Align | self-supervised articulated pose estimation · Synthetic dataset | Joint direction error (degree): 23.09 -> 6.63 | Improved |
| replace EAP with OP-Align | self-supervised articulated pose estimation · Synthetic dataset | Translation error: 0.121 -> 0.129 | Hurt |
| add shape variance regularization L_regS (ablation) | OP-Align · Real-world dataset (laptop) | Joint and part mAP | Improved |
| add density regularization L_regD (ablation) | OP-Align · Real-world dataset | overall mAP | Improved |
| add segmentation regularization L_regW (ablation) | OP-Align · Real-world dataset | joint parameter validity | Improved |
| add joint pivot regularization L_regJ (ablation) | OP-Align · Real-world dataset | joint and part pose mAP | Improved |
| self-supervised OP-Align vs supervised PointNet++ | OP-Align · Real-world dataset | Segmentation mAP IoU50%: 42.20 -> 50.42 | Improved |
| apply OP-Align to basket category with small movable handles | OP-Align · Real-world basket | qualitative segmentation | Hurt |
Closest papers by meaning
- Self-Supervised Category-Level Articulated Object Pose Estimation with Part-Level SE(3) Equivariance (ICLR 2023 · similarity 0.92)
- Category-Level Articulated Object Pose Estimation (CVPR 2020 · similarity 0.91)
- Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos (CVPR 2024 · similarity 0.89)
- EfficientCAPER: An End-to-End Framework for Fast and Robust Category-Level Articulated Object Pose Estimation (NeurIPS 2024 · similarity 0.89)
- Rethinking Correspondence-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.89)
- Emergence of a Shared Canonical Object Frame from In-the-Wild Videos (NeurIPS 2026 · similarity 0.89)
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.
