Research map / 3D vision / Human pose estimation
Robust Category-Level 6D Pose Estimation with Coarse-to-Fine Rendering of Neural Features
Wufei Ma, Angtian Wang, Alan Yuille, Adam Kortylewski
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 generative 6D proposals instead of Faster R-CNN proposals (ablation) | Ours · Occluded PASCAL3D+ level 3 (ImageNet) | Pose Acc (pi/6): 23.53 -> 27.43 | Improved |
| use generative 6D proposals instead of Faster R-CNN proposals (ablation) | Ours · Occluded PASCAL3D+ level 2 (ImageNet) | Pose Acc (pi/6): 45.43 -> 47.95 | Improved |
| use generative 6D proposals instead of Faster R-CNN proposals (ablation) | Ours · Occluded PASCAL3D+ level 1 (ImageNet) | Pose Acc (pi/6): 65.79 -> 66.63 | Improved |
| use generative 6D proposals instead of Faster R-CNN proposals (ablation) | Ours · Occluded PASCAL3D+ level 1 (ImageNet) | Pose Acc (pi/18): 34.56 -> 30.84 | Hurt |
| use generative 6D proposals instead of Faster R-CNN proposals (ablation) | Ours · Occluded PASCAL3D+ level 2 (ImageNet) | Pose Acc (pi/18): 17.46 -> 16.25 | Hurt |
| add multi-object reasoning (ablation) | Ours · PASCAL3D+ PASCAL VOC | Pose Acc (pi/6): 44.89 -> 45.32 | Improved |
| add multi-object reasoning (ablation) | Ours · PASCAL3D+ PASCAL VOC | mAP: 0.41 -> 0.43 | Improved |
| increase initial 3D locations from 3x3x3 to 9x9x9 (ablation) | Ours · PASCAL3D+ ImageNet | Pose Acc (pi/6): 69.77 -> 81.45 | Improved |
| increase initial 3D poses from 6x2x2 to 12x3x3 (ablation) | Ours · PASCAL3D+ ImageNet | Pose Acc (pi/6): 71.64 -> 81.45 | Improved |
| increase initial 3D poses from 12x3x3 to 18x6x6 (ablation) | Ours · PASCAL3D+ ImageNet | Pose Acc (pi/6): 81.45 -> 83.46 | Improved |
| increase initial 3D locations from 9x9x9 to 12x12x12 (ablation) | Ours · PASCAL3D+ ImageNet | Pose Acc (pi/6): 81.45 -> 84.54 | Improved |
Closest papers by meaning
- Category-Level 6D Object Pose Estimation in the Wild: A Semi-Supervised Learning Approach and A New Dataset (NeurIPS 2022 · similarity 0.89)
- FS6D: Few-Shot 6D Pose Estimation of Novel Objects (CVPR 2022 · similarity 0.89)
- ZebraPose: Coarse To Fine Surface Encoding for 6DoF Object Pose Estimation (CVPR 2022 · similarity 0.88)
- Occlusion-Aware Self-Supervised Monocular 6D Object Pose Estimation (TPAMI 2021 · similarity 0.88)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 2025 · similarity 0.87)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · 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.
