Research map / 3D vision / Human pose estimation
PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation
Sida Peng, Yuan Liu, Qixing Huang, Xiaowei Zhou, Hujun Bao
Check 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 direct keypoint-coordinate regression with pixel-wise voting (ablation) | Keypoint-based pose estimation · Occlusion LINEMOD | ADD(-S): 6.42 -> 33.88 | Improved |
| Select keypoints on the object surface using FPS rather than bounding-box corners (ablation) | PVNet · Occlusion LINEMOD | ADD(-S): 33.88 -> 39.76 | Improved |
| Increase the number of FPS surface keypoints from 4 to 8 (ablation) | PVNet · Occlusion LINEMOD | ADD(-S): 17.96 -> 39.76 | Improved |
| Increase the number of FPS surface keypoints from 8 to 12 (ablation) | PVNet · Occlusion LINEMOD | ADD(-S): 39.76 -> 39.92 | No clear effect |
| Use uncertainty-driven PnP instead of EPnP (ablation) | PVNet with 8 FPS keypoints · Occlusion LINEMOD | ADD(-S): 39.76 -> 40.77 | Improved |
| Use pixel-wise voting with RANSAC-based keypoint localization | PVNet · LINEMOD | ADD(-S): 55.95 -> 86.27 | Improved |
| Use PVNet for keypoint localization rather than the strongest compared method | Pose estimation · Occlusion LINEMOD | ADD(-S): 30.4 -> 40.77 | Improved |
| Use PVNet for pose estimation | Pose estimation · YCB-Video | ADD(-S) AUC: 72.8 -> 73.4 | Improved |
| Train and test the vector-field voting approach on cropped images with only 40% to 60% of the target visible | PVNet · Truncation LINEMOD | ADD(-S) | Mixed |
Later papers that compared with this method
17 later accepted papers report PVNet as a baseline in their results tables; 14 of them report a better number on most of the results they share with it.
- Prior-free 3D Object Tracking (CVPR 2025 · better on 1 of 1 shared result)
- OnePose: One-Shot Object Pose Estimation Without CAD Models (CVPR 2022 · better on 7 of 7 shared results)
- OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Models (NeurIPS 2022 · better on 1 of 3 shared results)
- Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images (ECCV 2022 · better on 3 of 3 shared results)
- DCL-Net: Deep Correspondence Learning Network for 6D Pose Estimation (ECCV 2022 · better on 1 of 1 shared result)
- RePOSE: Fast 6D Object Pose Refinement via Deep Texture Rendering (ICCV 2021 · better on 3 of 3 shared results)
- PR-GCN: A Deep Graph Convolutional Network With Point Refinement for 6D Pose Estimation (ICCV 2021 · better on 2 of 2 shared results)
- Keypoint-Graph-Driven Learning Framework for Object Pose Estimation (CVPR 2021 · better on 0 of 1 shared result)
9 more not listed.
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
- PVNet: Pixel-Wise Voting Network for 6DoF Object Pose Estimation (TPAMI 2020 · similarity 0.99)
- PVN3D: A Deep Point-Wise 3D Keypoints Voting Network for 6DoF Pose Estimation (CVPR 2020 · similarity 0.91)
- Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting (ECCV 2022 · similarity 0.90)
- CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation (ICCV 2019 · similarity 0.88)
- Segmentation-Driven 6D Object Pose Estimation (CVPR 2019 · similarity 0.87)
- FAR: Flexible Accurate and Robust 6DoF Relative Camera Pose Estimation (CVPR 2024 · 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.
