Research map / 3D vision / Human pose estimation
CoVisPose: Co-Visibility Pose Transformer for Wide-Baseline Relative Pose Estimation in 360° Indoor Panoramas
Will Hutchcroft, Yuguang Li, Ivaylo Boyadzhiev, Zhiqiang Wan, Haiyan Wang, Sing Bing Kang
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 co-visibility output to boundary-only training (ablation) | CoVisPose-Direct · ZInD test | rotation error mean (deg): 16.35 -> 4.94 | Improved |
| add co-visibility output to boundary-only training (ablation) | CoVisPose-Direct · ZInD test | translation error <0.5 m (%): 51.89 -> 86.13 | Improved |
| add angular correspondence output (ablation) | CoVisPose-Direct · ZInD test | rotation error mean (deg): 4.94 -> 4.33 | Improved |
| add angular correspondence output (ablation) | CoVisPose-Direct · ZInD test | rotation error <2.5 deg (%): 74.34 -> 80.17 | Improved |
| add angular correspondence output (ablation) | CoVisPose-Direct · ZInD test | translation error <0.5 m (%): 86.13 -> 88.72 | Improved |
| replace direct regression with RANSAC on predicted correspondences | CoVisPose · ZInD test (co-visibility 75-100%) | rotation error mean (deg): 1.27 -> 1.20 | Mixed |
| replace direct regression with RANSAC on predicted correspondences | CoVisPose · ZInD test (co-visibility 10-25%) | success rate (%): 100.00% -> 88.46% | Hurt |
| replace direct regression with RANSAC on predicted correspondences | CoVisPose · inference time | latency: 50 ms -> 5-30 secs | Hurt |
| use CoVisPose instead of DirectionNet at low overlap | CoVisPose-Direct · ZInD test (co-visibility 10-25%) | rotation error mean (deg): 53.37 -> 8.90 | Improved |
| use CoVisPose instead of LoFTR-OpenMVG | CoVisPose-Direct · ZInD test (co-visibility 75-100%) | rotation error mean (deg): 15.71 -> 1.27 | Improved |
Closest papers by meaning
- Alligat0R: Pre-Training through Covisibility Segmentation for Relative Camera Pose Regression (NeurIPS 2025 · similarity 0.87)
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.86)
- CM-EVS: Sparse Panoramic RGB-D-Pose Data for Complete Scene Coverage (NeurIPS 2026 · similarity 0.86)
- GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024 · similarity 0.85)
- SparseGNV: Generating Novel Views of Indoor Scenes with Sparse RGB-D Images (AAAI 2024 · similarity 0.85)
- Co-op: Correspondence-based Novel Object Pose Estimation (CVPR 2025 · 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.
