Research map / 3D vision / Human pose estimation
Neural Reprojection Error: Merging Feature Learning and Camera Pose Estimation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 RE-based estimation with NRE while using the same S2DNet features | S2DNet pose estimation · MegaDepth | translation error at 0.25m: 0.51 -> 0.44 | Helped |
| Replace RE-based estimation with NRE while using the same S2DNet features | S2DNet pose estimation · MegaDepth | rotation error at 2°: 0.51 -> 0.44 | Helped |
| Train features with NRE instead of using S2DNet features | NRE pose estimator · MegaDepth Easy | translation error at 0.25m: 0.17 -> 0.12 | Helped |
| Train features with NRE instead of using S2DNet features | NRE pose estimator · MegaDepth Medium | translation error at 0.25m: 0.29 -> 0.19 | Helped |
| Train features with NRE instead of using S2DNet features | NRE pose estimator · MegaDepth Hard | translation error at 0.25m: 0.44 -> 0.34 | Helped |
| Use NRE pose estimation instead of FPR | S2DNet pose estimation · Aachen Night | pose accuracy at 0.25m, 2°: 0.32 -> 0.30 | Hurt |
| Use NRE pose estimation instead of RE Min. Eq. 10 | S2DNet pose estimation · InLoc-DUC1 | pose accuracy at 0.25m, 2°: 0.58 -> 0.60 | Helped |
| Add successive steps of the coarse-to-fine estimator (ablation) | NRE pose estimator · MegaDepth Hard | pose estimation error | Helped |
Closest papers by meaning
- LEAP: Liberate Sparse-View 3D Modeling from Camera Poses (ICLR 2024 · similarity 0.89)
- FAR: Flexible Accurate and Robust 6DoF Relative Camera Pose Estimation (CVPR 2024 · similarity 0.88)
- Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization (CVPR 2025 · similarity 0.88)
- Sparse-view Pose Estimation and Reconstruction via Analysis by Generative Synthesis (NeurIPS 2024 · similarity 0.88)
- PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization (CVPR 2024 · similarity 0.88)
- Neural Refinement for Absolute Pose Regression with Feature Synthesis (CVPR 2024 · similarity 0.88)
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.