Research map / 3D vision / Human pose estimation
RayPose: Ray Bundling Diffusion for Template Views in Unseen 6D Object Pose Estimation
Junwen Huang, Shishir Reddy Vutukur, Peter KT Yu, Nassir Navab, Slobodan Ilic, Benjamin Busam
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 fine predictor with narrowed template sampling (ablation) | RayPose coarse · LM-O (ablation) | AR: 39.6 -> 42.1 | Improved |
| add external MegaPose refinement to fine predictor (ablation) | RayPose Fine · LM-O (ablation) | AR: 42.1 -> 56.2 | Improved |
| predict relative pose to templates instead of absolute pose (ablation) | RayPose · LM-O (ablation) | AR: 35.4 -> 39.6 | Improved |
| use multiple posed templates instead of single view (ablation) | RayPose · LM-O (ablation) | AR: 32.5 -> 35.4 | Improved |
| input ground-truth template pose maps (ablation) | RayPose · LM-O (ablation) | AR: 27.8 -> 39.6 | Improved |
| random template sampling instead of fixed templates for coarse predictor (ablation) | RayPose Coarse · LM-O (template distribution) | AR: 60.28 -> 65.58 | Improved |
| narrow template range to +-30deg, +-5cm (ablation) | RayPose Fine · LM-O (template distribution) | AR: 65.81 -> 67.29 | Improved |
| over-narrow template range to +-15deg, +-3cm (ablation) | RayPose Fine · LM-O (template distribution) | AR: 67.29 -> 61.04 | Hurt |
| wide range +-90deg, +-10cm versus random coarse sampling (ablation) | RayPose Fine · LM-O (template distribution) | AR: 65.58 -> 65.81 | No clear effect |
| replace FoundPose with RayPose, no refinement | FoundPose · BOP 5-dataset average | AR: 37.8 -> 39.1 | Improved |
| replace FoundPose with RayPose on IC-BIN, no refinement | FoundPose · BOP IC-BIN | AR: 23.9 -> 21.8 | Hurt |
| replace GigaPose with RayPose on IC-BIN, with refinement | GigaPose · BOP IC-BIN | AR: 44.3 -> 41.6 | Hurt |
Later papers that compared with this method
1 later accepted paper reports RayPose as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- PoseGAM: Robust Unseen Object Pose Estimation via Geometry-Aware Multi-View Reasoning (CVPR 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GigaPose — GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence (CVPR 2024) · better on 3 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
- Cameras as Rays: Pose Estimation via Ray Diffusion (ICLR 2024 · similarity 0.90)
- Co-op: Correspondence-based Novel Object Pose Estimation (CVPR 2025 · similarity 0.88)
- Object Pose Estimation via the Aggregation of Diffusion Features (CVPR 2024 · similarity 0.88)
- RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objects (CVPR 2025 · similarity 0.87)
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.87)
- FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (CVPR 2024 · similarity 0.87)
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.
