Research map / 3D vision / Human pose estimation
DProST: Dynamic Projective Spatial Transformer Network for 6D Pose Estimation
Jaewoo Park, Nam Ik Cho
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 a second refinement iteration (1 -> 2) (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 22.6 -> 48.1 | Improved |
| add a third refinement iteration (2 -> 3) (ablation) | DProST · LINEMOD | ADD(-S) 0.10d: 97.7 -> 97.7 | No clear effect |
| add a third refinement iteration (2 -> 3) (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 48.1 -> 50.8 | Improved |
| add GD loss to GM loss (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 35.5 -> 48.1 | Improved |
| add GD loss to GM loss (ablation) | DProST · LINEMOD | ADD(-S) 0.10d: 94.5 -> 97.7 | Improved |
| add GD loss to GM loss (ablation) | DProST · LINEMOD | Proj2D 2pix: 86.6 -> 89.3 | Improved |
| use 64 instead of 32 points per ray (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 46.3 -> 48.1 | Improved |
| use 128 instead of 64 points per ray (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 48.1 -> 47.1 | Hurt |
| replace the space-carved reference with the mesh (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 48.1 -> 51.6 | Improved |
| use FPS instead of random reference view sampling (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 45.9 -> 48.1 | Improved |
| use 8 instead of 4 reference images (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 44.1 -> 48.1 | Improved |
| use 16 instead of 8 reference images (ablation) | DProST · LINEMOD | ADD(-S) 0.02d: 48.1 -> 45.6 | Hurt |
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.
- GDR-Net — GDR-Net: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation (CVPR 2021) · 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
- Real-Time Seamless Single Shot 6D Object Pose Prediction (CVPR 2018 · similarity 0.88)
- GDRNPP: A Geometry-Guided and Fully Learning-Based Object Pose Estimator (TPAMI 2025 · similarity 0.87)
- Segmentation-Driven 6D Object Pose Estimation (CVPR 2019 · similarity 0.87)
- Reconstruct Locally, Localize Globally: A Model Free Method for Object Pose Estimation (CVPR 2020 · similarity 0.87)
- CatFormer: Category-Level 6D Object Pose Estimation with Transformer (AAAI 2024 · similarity 0.86)
- 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.
