Research map / Detection and segmentation / Open-set and few-shot detection
DecomPose: Disentangling Cross-Category Optimization Contention for Category-Level 6D Object Pose Estimation
Yifan Gao, Lu Zou, Zhangjin Huang, Guoping Wang
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 difficulty-aware grouping instead of random grouping (ablation) | DecomPose · REAL275 | 5deg2cm mAP: 59.2 -> 61.1 | Improved |
| use difficulty-aware grouping instead of random grouping (ablation) | DecomPose · REAL275 | 5deg5cm mAP: 66.4 -> 67.9 | Improved |
| use difficulty-aware grouping instead of random grouping (ablation) | DecomPose · HouseCat6D | 5deg2cm mAP: 20.7 -> 25.3 | Improved |
| add random grouping (ablation) | AG-Pose · HouseCat6D | 5deg2cm mAP: 21.3 -> 20.7 | Hurt |
| use asymmetric branches (lightweight branch for hardest group) instead of uniform heavy (ablation) | DecomPose · REAL275 | 5deg2cm mAP: 59.4 -> 61.1 | Improved |
| use asymmetric branches instead of uniform heavy (ablation) | DecomPose · REAL275 | IoU75: 80.6 -> 81.1 | Improved |
| use asymmetric H/H/L/L instead of uniform H/H/H/H (ablation) | DecomPose · HouseCat6D | 5deg2cm mAP: 23.1 -> 25.3 | Improved |
| decompose correspondence module instead of backbone or pose head (ablation) | DecomPose · REAL275 | 5deg2cm mAP: 57.3 -> 61.1 | Improved |
| increase number of groups from 3 to 4 (ablation) | DecomPose · REAL275 | 5deg2cm mAP: 61.1 -> 60.5 | Hurt |
| increase number of groups from 4 to 5 (ablation) | DecomPose · HouseCat6D | 5deg2cm mAP: 25.3 -> 23.8 | Hurt |
| change reference estimator for difficulty scores (IST-Net, SecondPose, AG-Pose) (ablation) | DecomPose · REAL275 | 5deg2cm mAP: 61.3, 61.0, 61.1 | No clear effect |
| add difficulty-aware grouping on heterogeneous HouseCat6D subset (ablation) | DecomPose · HouseCat6D (6-category subset) | 5deg2cm mAP: 10.3 -> 11.9 | Improved |
Closest papers by meaning
- FS-Net: Fast Shape-Based Network for Category-Level 6D Object Pose Estimation With Decoupled Rotation Mechanism (CVPR 2021 · similarity 0.88)
- PriorPose: Reference-Guided Joint Deformation and Alignment for Category-Level Object Pose Estimation (ECCV 2026 · similarity 0.88)
- Joint Learning of Pose Regression and Denoising Diffusion with Score Scaling Sampling for Category-level 6D Pose Estimation (ICCV 2025 · similarity 0.88)
- GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement (CVPR 2024 · similarity 0.86)
- One2Any: One-Reference 6D Pose Estimation for Any Object (CVPR 2025 · similarity 0.86)
- GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.86)
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.
