Research map / 3D vision / Human pose estimation
DPODv2: Dense Correspondence-Based 6 DoF Pose Estimation
Ivan Shugurov, Sergey Zakharov, Slobodan Ilić
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 4 calibrated-view refinement | DPODv2 RGB · Linemod (real train) | ADD recall: 93.59 -> 99.91 | Improved |
| add 4 calibrated-view refinement | DPODv2 RGB · Linemod (synthetic train) | ADD recall: 81.20 -> 99.70 | Improved |
| add 4 calibrated-view refinement | DPODv2 RGB · Homebrewed | average recall: 0.725 -> 0.835 | Improved |
| add ICP refinement | DPODv2 + Kabsch · TLESS (PBR train) | average recall: 0.699 -> 0.590 | Hurt |
| add ICP refinement | DPODv2 + Kabsch · Occlusion (PBR train) | average recall: 0.700 -> 0.698 | No clear effect |
| use depth CENet instead of RGB CENet | DPODv2 · Homebrewed | average recall: 0.725 -> 0.382 | Hurt |
| use RGB correspondences with depth Kabsch instead of RGB-only PnP | DPODv2 · Homebrewed | average recall: 0.725 -> 0.84 | Improved |
| train correspondence network on depth instead of RGB | DPODv2 · Linemod (synthetic train) | ADD recall: 99.22 -> 88.94 | Hurt |
| choose furthest instead of closest views for refinement (2 views) (ablation) | multi-view refiner · Linemod (real train) | ADD recall: 86.43 -> 99.71 | Improved |
| choose random instead of closest views for refinement (2 views) (ablation) | multi-view refiner · Homebrewed (synthetic) | average recall: 0.693 -> 0.808 | Improved |
| choose furthest instead of closest views (4 views) (ablation) | multi-view refiner · Linemod (real train) | ADD recall: 96.53 -> 99.97 | Improved |
| train correspondence network on synthetic instead of real data (GT patches) | DPODv2 · TLESS | average recall: 0.535 -> 0.729 | Improved |
Later papers that compared with this method
1 later accepted paper reports DpodV2 as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- WeLSA: Learning to Predict 6D Pose from Weakly Labeled Data Using Shape Alignment (ECCV 2022 · better on 0 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 on most shared results.
- DPOD — DPOD: 6D Pose Object Detector and Refiner (ICCV 2019) · better on 2 of 2 shared results
- PVN3D — PVN3D: A Deep Point-Wise 3D Keypoints Voting Network for 6DoF Pose Estimation (CVPR 2020) · better on 1 of 1 shared result
- PVNet — PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation (CVPR 2019) · better on 1 of 1 shared result
- Pix2Pose — Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation (ICCV 2019) · better on 1 of 1 shared result
- SSD6D — SSD-6D: Making RGB-Based 3D Detection and 6D Pose Estimation Great Again (ICCV 2017) · better on 1 of 1 shared result
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
- DPOD: 6D Pose Object Detector and Refiner (ICCV 2019 · similarity 0.94)
- OSOP: A Multi-Stage One Shot Object Pose Estimation Framework (CVPR 2022 · similarity 0.88)
- SingRef6D: Monocular Novel Object Pose Estimation with a Single RGB Reference (NeurIPS 2025 · similarity 0.87)
- Co-op: Correspondence-based Novel Object Pose Estimation (CVPR 2025 · similarity 0.86)
- MatchU: Matching Unseen Objects for 6D Pose Estimation from RGB-D Images (CVPR 2024 · similarity 0.86)
- HiPose: Hierarchical Binary Surface Encoding and Correspondence Pruning for RGB-D 6DoF Object Pose Estimation (CVPR 2024 · 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.
