Research map / 3D vision / Human pose estimation
Learning Symmetry-Aware Geometry Correspondences for 6D Object Pose Estimation
Heng Zhao, Shenxing Wei, Dahu Shi, Wenming Tan, Zheyang Li, Ye Ren, Xing Wei, Yi Yang, Shiliang Pu
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 GCPose instead of OVE6D without ICP | unseen-object 6D pose estimation · T-LESS | AR: 0.523 -> 0.679 | Improved |
| use GCPose instead of OVE6D without ICP | unseen-object 6D pose estimation · LINEMOD | AR: 0.701 -> 0.841 | Improved |
| use GCPose without ICP versus OVE6D with ICP | unseen-object 6D pose estimation · LINEMOD | AR: 0.835 -> 0.841 | Improved |
| add special data augmentation for synthetic depth training (ablation) | GCPose · T-LESS (GT label and mask) | AR: 0.635 -> 0.669 | Improved |
| add point cloud normalization by model diameter (ablation) | GCPose · T-LESS (GT label and mask) | AR: 0.669 -> 0.698 | Improved |
| add global-aware fine-level features (ablation) | GCPose · T-LESS (GT label and mask) | AR: 0.698 -> 0.708 | Improved |
| add symmetry-aware matching loss (ablation) | GCPose · T-LESS (GT label and mask) | AR: 0.708 -> 0.738 | Improved |
| scale training data by adding 100k ABC CAD models to ShapeNet (ablation) | GCPose · T-LESS | AR: 0.679 -> 0.724 | Improved |
| scale training data with ABC on TUD-L (ablation) | GCPose · TUD-L | AR: 0.926 -> 0.942 | Improved |
| use ground-truth label and mask instead of Mask R-CNN segmentation | GCPose · T-LESS | AR: 0.679 -> 0.738 | Improved |
Closest papers by meaning
- Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World (ECCV 2022 · similarity 0.91)
- Match Normalization: Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World (TPAMI 2024 · similarity 0.90)
- RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objects (CVPR 2025 · similarity 0.87)
- Instance-Adaptive and Geometric-Aware Keypoint Learning for Category-Level 6D Object Pose Estimation (CVPR 2024 · similarity 0.87)
- Structure-Aware Correspondence Learning for Relative Pose Estimation (CVPR 2025 · similarity 0.87)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 2025 · 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.
