Research map / 3D vision / Human pose estimation
ComPose: A Unified Completion-Pose Framework for Robust Category-Level Object Pose Estimation
Huan Ren, Yihan Chen, Chuxin Wang, Nailong Liu, Wenfei Yang, Tianzhu Zhang
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 |
|---|---|---|---|
| replace complete-shape recovery with partial instance reconstruction (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 55.6 -> 49.6 | Hurt |
| add completed dense point cloud to enhance keypoints (ablation) | ComPose · REAL275 depth-only | 10deg5cm: 83.3 -> 85.0 | Improved |
| use progressive completion instead of AdaPoinTr (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 53.4 -> 55.6 | Improved |
| use progressive completion instead of PoinTr (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 52.7 -> 55.6 | Improved |
| use progressive completion instead of static query (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 51.6 -> 55.6 | Improved |
| add adaptive keypoint selection (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 53.8 -> 55.6 | Improved |
| add geometric relation encoding (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 49.5 -> 53.8 | Improved |
| add geometric relation consistency constraint (ablation) | ComPose · REAL275 depth-only | 5deg2cm: 53.8 -> 55.6 | Improved |
| use unified completion-pose instead of two-stage completion-then-pose (ablation) | ComPose · REAL275 depth-only | accuracy and latency | Improved |
| apply 25% occlusion augmentation at test time (ablation) | AG-Pose* · REAL275 depth-only | 5deg5cm drop: 16.5% | Hurt |
| apply 25% occlusion augmentation at test time (ablation) | ComPose · REAL275 depth-only | 5deg5cm drop: 12.6% | Hurt |
| use ComPose instead of AG-Pose in depth-only setting | pose estimator · REAL275 depth-only | 5deg2cm: 48.8 -> 55.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- GCE-Pose — GCE-Pose: Global Context Enhancement for Category-level Object Pose Estimation (CVPR 2025) · better on 4 of 4 shared results
- CleanPose — CleanPose: Category-Level Object Pose Estimation via Causal Learning and Knowledge Distillation (ICCV 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement (CVPR 2024 · similarity 0.89)
- Learning Shape-Independent Transformation via Spherical Representations for Category-Level Object Pose Estimation (ICLR 2025 · similarity 0.89)
- Rethinking Correspondence-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.89)
- SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation (CVPR 2024 · similarity 0.89)
- Point-based Instance Completion with Scene Constraints (ICLR 2025 · similarity 0.89)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · 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.
