Research map / 3D vision / Human pose estimation
OnePose: One-Shot Object Pose Estimation Without CAD Models
Check 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 |
|---|---|---|---|
| Directly match query 2D features to the SfM map using OnePose rather than HLoc (SPP + SPG) | OnePose vs. HLoc · OnePose dataset, large objects, 1cm-1deg | pose accuracy: 0.435 -> 0.471 | Helped |
| Directly match query 2D features to the SfM map using OnePose rather than HLoc (SPP + SPG) | OnePose vs. HLoc · OnePose dataset, medium objects, 1cm-1deg | pose accuracy: 0.643 -> 0.629 | Hurt |
| Directly match query 2D features to the SfM map using OnePose rather than HLoc (SPP + SPG) | OnePose vs. HLoc · OnePose dataset, small objects, 1cm-1deg | pose accuracy: 0.432 -> 0.405 | Hurt |
| Use OnePose's direct 2D-3D matching instead of HLoc (SPP + SPG) | OnePose vs. HLoc · OnePose dataset | runtime (ms): 618.29 -> 58.31 | Helped |
| Use OnePose rather than PVNet | OnePose · OnePose dataset, selected objects, 5cm-5deg | pose accuracy | Helped |
| Use OnePose rather than Objectron | OnePose · OnePose dataset, Shoe and Cup objects | mean pixel error of 2D projection | Helped |
| Add adaptive aggregation-attention over feature tracks instead of averaging descriptors (ablation) | OnePose · OnePose dataset, large objects, 1cm-1deg | pose accuracy: 0.449 -> 0.471 | Helped |
| Add adaptive aggregation-attention over feature tracks instead of averaging descriptors (ablation) | OnePose · OnePose dataset, medium objects, 1cm-1deg | pose accuracy: 0.602 -> 0.629 | Helped |
| Add adaptive aggregation-attention over feature tracks instead of averaging descriptors (ablation) | OnePose · OnePose dataset, small objects, 1cm-1deg | pose accuracy: 0.390 -> 0.405 | Helped |
| Use GNN matching instead of nearest-neighbor matching with average aggregation (ablation) | OnePose · OnePose dataset, large objects, 1cm-1deg | pose accuracy: 0.426 -> 0.449 | Helped |
| Use GNN matching instead of nearest-neighbor matching with average aggregation (ablation) | OnePose · OnePose dataset, medium objects, 1cm-1deg | pose accuracy: 0.570 -> 0.602 | Helped |
| Use K-Means aggregation instead of average aggregation with NN matching (ablation) | OnePose · OnePose dataset, large objects, 1cm-1deg | pose accuracy: 0.426 -> 0.431 | Helped |
Closest papers by meaning
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.88)
- One2Any: One-Reference 6D Pose Estimation for Any Object (CVPR 2025 · similarity 0.87)
- MFOS: Model-Free & One-Shot Object Pose Estimation (AAAI 2024 · similarity 0.86)
- FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (CVPR 2024 · similarity 0.86)
- NOPE: Novel Object Pose Estimation from a Single Image (CVPR 2024 · similarity 0.86)
- ProbPose: A Probabilistic Approach to 2D Human Pose Estimation (CVPR 2025 · similarity 0.85)
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.