Research map / 3D vision / Human pose estimation
OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Models
Xingyi He, Jiaming Sun, Yuang Wang, Di Huang, Hujun Bao, Xiaowei Zhou
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 keypoint-free semi-dense reconstruction and direct sparse-to-dense 2D-3D matching | OnePose++ · OnePose-LowTexture | 3cm-3deg pose success rate: 35.7 -> 57.7 | Improved |
| use keypoint-free semi-dense reconstruction and direct sparse-to-dense 2D-3D matching | OnePose++ · LINEMOD | ADD(S)-0.1d: 63.6 -> 76.9 | Improved |
| use keypoint-free semi-dense reconstruction and direct sparse-to-dense 2D-3D matching | OnePose++ · OnePose-LowTexture (objects with scanned models) | ADD(S)-0.1d: 62.7 -> 88.9 | Improved |
| include point-cloud refinement rather than using coarse reconstruction only (ablation) | OnePose++ · OnePose | 1cm-1deg pose success rate: 43.9 -> 51.1 | Improved |
| include point-cloud refinement rather than using coarse reconstruction only (ablation) | OnePose++ · OnePose-LowTexture point cloud | 1mm accuracy: 26.6 -> 30.9 | Improved |
| include the 2D-3D attention module instead of directly matching backbone features (ablation) | OnePose++ · OnePose-LowTexture | 5cm-5deg pose success rate: 55.3 -> 72.1 | Improved |
| use keypoint-free reconstruction and matching instead of OnePose's keypoint-based approach | OnePose++ · OnePose | 1cm-1deg pose success rate: 49.7 -> 51.1 | No clear effect |
| use keypoint-free reconstruction and matching instead of OnePose's keypoint-based approach | OnePose++ · LINEMOD | Proj2D: 78.1 -> 94.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- OnePose — OnePose: One-Shot Object Pose Estimation Without CAD Models (CVPR 2022) · better on 5 of 5 shared results
- PVNet — PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation (CVPR 2019) · better on 1 of 3 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
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.89)
- One2Any: One-Reference 6D Pose Estimation for Any Object (CVPR 2025 · similarity 0.87)
- Rethinking Correspondence-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.87)
- KeyPose: Category-Level 6D Object Pose Estimation with Self-Adaptive Keypoints (AAAI 2025 · similarity 0.87)
- NOPE: Novel Object Pose Estimation from a Single Image (CVPR 2024 · similarity 0.86)
- Structure-Aware Correspondence Learning for Relative 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.
