Research map / 3D vision / Human pose estimation
Category-Level 6D Object Pose Estimation in the Wild: A Semi-Supervised Learning Approach and A New Dataset
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 |
|---|---|---|---|
| Train with unlabeled Wild6D data rather than CAMERA25 alone (ablation) | RePoNet-semi · REAL275 | IOU@0.75: 45.7 -> 52.5 | Helped |
| Train with unlabeled REAL275 data rather than CAMERA25 alone (ablation) | RePoNet-semi · REAL275 | IOU@0.75: 45.7 -> 49.2 | Helped |
| Add both unlabeled REAL275 and Wild3D data (ablation) | RePoNet-semi · REAL275 | IOU@0.5: 83.8 -> 86.0 | Helped |
| Add an intermediate NOCS map (ablation) | RePoNet · REAL275, fully supervised | IOU@0.5: 77.0 -> 79.1 | Helped |
| Add an implicit function (ablation) | RePoNet · REAL275, fully supervised | IOU@0.5: 77.0 -> 79.2 | Helped |
| Add both the intermediate NOCS map and implicit function (ablation) | RePoNet · REAL275, fully supervised | IOU@0.5: 77.0 -> 81.1 | Helped |
| Add silhouette matching loss and pose regression loss (ablation) | RePoNet · REAL275, fully supervised | IOU@0.5: 78.1 -> 81.1 | Helped |
| Add silhouette matching loss (ablation) | RePoNet · REAL275, fully supervised | IOU@0.75: 61.9 -> 63.7 | Helped |
| Use Wild6D rather than CO3D as the added unlabeled video dataset (ablation) | RePoNet · REAL275, bottle | 5 degree, 5cm: 45.8 -> 61.1 | Helped |
| Use CO3D data in addition to CAMERA25 (ablation) | RePoNet · REAL275, bottle | 5 degree, 5cm: 43.1 -> 45.8 | Helped |
| Train on CAMERA75 and Wild6D instead of CAMERA75 alone | RePoNet · Wild6D test set | 5 degree, 2cm: 26.0 -> 29.5 | Helped |
| Train on CAMERA75 and Wild6D instead of CAMERA75 alone | RePoNet · Wild6D test set | 5 degree, 5cm: 30.8 -> 34.4 | Helped |
Closest papers by meaning
- RGBD Objects in the Wild: Scaling Real-World 3D Object Learning from RGB-D Videos (CVPR 2024 · similarity 0.89)
- 6D Object Pose Tracking in Internet Videos for Robotic Manipulation (ICLR 2025 · similarity 0.89)
- Any6D: Model-free 6D Pose Estimation of Novel Objects (CVPR 2025 · similarity 0.88)
- Open-Vocabulary Object 6D Pose Estimation (CVPR 2024 · similarity 0.88)
- FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (CVPR 2024 · similarity 0.88)
- SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimation (CVPR 2024 · similarity 0.88)
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.