Research map / 3D vision / Human pose estimation
RePOSE: Fast 6D Object Pose Refinement via Deep Texture Rendering
Shun Iwase, Xingyu Liu, Rawal Khirodkar, Rio Yokota, Kris M. Kitani
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 learnable deep textures with RGB photometric error (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 96.1 -> 23.0 | Improved |
| replace learnable deep textures with RGB photometric error (ablation) | RePOSE · Occlusion LineMOD ADD(-S) | ADD(-S): 51.6 -> 13.1 | Improved |
| use deep texture rendering instead of CNN features with feature warping (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 90.7 -> 96.1 | Improved |
| use deep texture rendering instead of CNN features with feature warping (ablation) | RePOSE · Occlusion LineMOD ADD(-S) | ADD(-S): 42.9 -> 51.6 | Improved |
| re-render deep texture each iteration instead of warping first-step features (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 95.5 -> 96.1 | Improved |
| re-render deep texture each iteration instead of warping first-step features (ablation) | RePOSE · Occlusion LineMOD ADD(-S) | ADD(-S): 44.5 -> 51.6 | Improved |
| use geometric LM refinement over deep textures instead of DPOD direct-regression refinement (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 90.5 -> 96.1 | Improved |
| use geometric LM refinement over deep textures instead of DPOD direct-regression refinement (ablation) | RePOSE · Occlusion LineMOD ADD(-S) | ADD(-S): 42.6 -> 51.6 | Improved |
| increase LM refinement iterations from 0 to 5 (ablation) | RePOSE · Occlusion LineMOD ADD(-S) | ADD(-S): 40.8 -> 51.6 | Improved |
| increase iterations from 1 to 5 (runtime cost) (ablation) | RePOSE · Occlusion LineMOD refinement runtime | runtime: 4.1 ms -> 10.9 ms | Hurt |
| set neural texture channels to 3 instead of 1 (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 95.2 -> 96.1 | Improved |
| increase neural texture channels from 3 to 9 (ablation) | RePOSE · LineMOD ADD(-S) | ADD(-S): 96.1 -> 94.5 | Hurt |
Later papers that compared with this method
2 later accepted papers report RePOSE as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- RNNPose: Recurrent 6-DoF Object Pose Refinement With Robust Correspondence Field Estimation and Pose Optimization (CVPR 2022 · better on 3 of 3 shared results)
- ZebraPose: Coarse To Fine Surface Encoding for 6DoF Object Pose Estimation (CVPR 2022 · better on 1 of 1 shared result)
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.
- PVNet — PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation (CVPR 2019) · better on 3 of 3 shared results
- DPOD — DPOD: 6D Pose Object Detector and Refiner (ICCV 2019) · 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
- RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objects (CVPR 2025 · similarity 0.89)
- SingRef6D: Monocular Novel Object Pose Estimation with a Single RGB Reference (NeurIPS 2025 · similarity 0.88)
- GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement (CVPR 2024 · similarity 0.87)
- The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement (CVPR 2024 · similarity 0.87)
- HiPose: Hierarchical Binary Surface Encoding and Correspondence Pruning for RGB-D 6DoF Object Pose Estimation (CVPR 2024 · similarity 0.87)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 2025 · 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.
