Research map / 3D vision / Human pose estimation
UF0-6D: Unified Flow-based Zero-Shot 6D Object Pose Estimation without Refinement
Yingnan Guo, Chun Lim, Yu Feng, Yu 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 |
|---|---|---|---|
| add query features F^q to conditioning (ablation) | UF0-6D (RGB) · BOP mean | AR: 55.9 -> 70.6 | Improved |
| add object/template features F^o to conditioning (ablation) | UF0-6D (RGB) · BOP mean | AR: 60.8 -> 70.6 | Improved |
| add both query and object features (ablation) | UF0-6D (RGB) · BOP mean | AR: 42.7 -> 70.6 | Improved |
| add query features F^q to conditioning (ablation) | UF0-6D (RGB-D) · BOP mean | AR: 69.1 -> 81.2 | Improved |
| add object/template features F^o to conditioning (ablation) | UF0-6D (RGB-D) · BOP mean | AR: 72.9 -> 81.2 | Improved |
| add both query and object features (ablation) | UF0-6D (RGB-D) · BOP mean | AR: 55.4 -> 81.2 | Improved |
| increase ODE steps from 1 to 20 (ablation) | UF0-6D (RGB) · LM-O | AR: 57.8 -> 66.2 | Improved |
| increase ODE steps from 1 to 20 (ablation) | UF0-6D (RGB-D) · LM-O | AR: 66.4 -> 75.9 | Improved |
| increase ODE steps from 20 to 40 (ablation) | UF0-6D (RGB) · LM-O | AR: 66.2 -> 66.3 | No clear effect |
| increase ODE steps from 20 to 40 (ablation) | UF0-6D (RGB-D) · LM-O | AR: 75.9 -> 76.0 | No clear effect |
| increase ODE steps from 20 to 40 (ablation) | UF0-6D (RGB) · LM-O | time (s): 0.379 -> 0.637 | Hurt |
| increase reference views from 3 to 12 (ablation) | UF0-6D · YCB-Video model-free | ADD-S: 93.8 -> 97.6 | Improved |
Closest papers by meaning
- Any6D: Model-free 6D Pose Estimation of Novel Objects (CVPR 2025 · similarity 0.89)
- SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimation (CVPR 2024 · similarity 0.89)
- SingRef6D: Monocular Novel Object Pose Estimation with a Single RGB Reference (NeurIPS 2025 · similarity 0.89)
- GenFlow: Generalizable Recurrent Flow for 6D Pose Refinement of Novel Objects (CVPR 2024 · similarity 0.89)
- One2Any: One-Reference 6D Pose Estimation for Any Object (CVPR 2025 · similarity 0.88)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 2025 · 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.
