Research map / Detection and segmentation / Open-set and few-shot detection
Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation
He Wang, Srinath Sridhar, Jingwei Huang, Julien Valentin, Shuran Song, Leonidas J. Guibas
Check 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 |
|---|---|---|---|
| predict dense NOCS maps and fit them to the depth point cloud instead of aligning a masked point cloud to a category model | 6D pose and size estimation · REAL275 | mAP 3D IoU@50%: 43.8% -> 76.4% | Improved |
| predict dense NOCS maps and fit them to the depth point cloud instead of aligning a masked point cloud to a category model | 6D pose and size estimation · REAL275 | mAP (5°, 5 cm): 0.8% -> 10.2% | Improved |
| add real-world training data and COCO images to CAMERA* training (ablation) | NOCS regression network · REAL275 | mAP 3D IoU@50%: 36.7 -> 72.4 | Improved |
| add real-world training data and COCO images to CAMERA* training (ablation) | NOCS regression network · REAL275 | mAP (5°, 5 cm): 3.4 -> 8.1 | Improved |
| add COCO to CAMERA* training data (ablation) | NOCS regression network · REAL275 | mAP (5°, 5 cm): 3.4 -> 3.3 | No clear effect |
| replace context-aware CAMERA* compositing with non-context-aware compositing (ablation) | NOCS regression network · REAL275 | mAP 3D IoU@50%: 72.4 -> 71.7 | Hurt |
| replace context-aware CAMERA* compositing with non-context-aware compositing (ablation) | NOCS regression network · REAL275 | mAP (5°, 5 cm): 8.1 -> 7.9 | Hurt |
| use pixel classification instead of regression for NOCS prediction (ablation) | NOCS prediction network · CAMERA25 | mAP (5°, 5 cm): 29.2 -> 40.9 | Improved |
| use pixel classification instead of regression for NOCS prediction (ablation) | NOCS prediction network · REAL275 | mAP (5°, 5 cm): 8.1 -> 10.0 | Improved |
| use 128 bins instead of 32 bins (ablation) | NOCS classification network · CAMERA25 | mAP 3D IoU@50%: 83.9 -> 85.3 | Improved |
| use 128 bins instead of 32 bins (ablation) | NOCS classification network · CAMERA25 | mAP (5°, 5 cm): 40.9 -> 38.8 | Hurt |
| use 128 bins instead of 32 bins (ablation) | NOCS classification network · REAL275 | mAP 3D IoU@50%: 78.0 -> 80.5 | Improved |
Closest papers by meaning
- Any6D: Model-free 6D Pose Estimation of Novel Objects (CVPR 2025 · similarity 0.89)
- SingRef6D: Monocular Novel Object Pose Estimation with a Single RGB Reference (NeurIPS 2025 · similarity 0.89)
- Open-Vocabulary Object 6D Pose Estimation (CVPR 2024 · similarity 0.89)
- GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.89)
- One2Any: One-Reference 6D Pose Estimation for Any Object (CVPR 2025 · similarity 0.89)
- CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image (CVPR 2025 · 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.
