Research map / 3D vision / Human pose estimation
SC-wLS: Towards Interpretable Feed-Forward Camera Re-localization
Xin Wu, Hao Zhao, Shunkai Li, Yingdian Cao, Hongbin Zha
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 end-to-end fine-tuning stage (ablation) | SC-wLS (dlt) · 7Scenes Stairs | median translation error (m): 0.179 -> 0.123 | Improved |
| add end-to-end fine-tuning stage (ablation) | SC-wLS (dlt) · Cambridge King's College | median translation error (m): 0.22 -> 0.14 | Improved |
| add end-to-end fine-tuning stage (ablation) | SC-wLS (dlt) · 7Scenes Chess | median translation error (m): 0.029 -> 0.029 | No clear effect |
| add self-supervised test-time adaptation (ablation) | SC-wLS (dlt+e2e) · 7Scenes Stairs | median translation error (m): 0.123 -> 0.055 | Improved |
| add self-supervised test-time adaptation (ablation) | SC-wLS (dlt+e2e) · Cambridge Greatcourt | median translation error (m): 1.64 -> 0.94 | Improved |
| add self-supervised test-time adaptation (ablation) | SC-wLS (dlt+e2e) · Cambridge Church | median translation error (m): 0.39 -> 0.17 | Improved |
| add end-to-end stage to LM-Refine pipeline (ablation) | SC-wLS (dlt+ref) · Cambridge King's College | median translation error (m): 0.09 -> 0.08 | Improved |
| add end-to-end stage to LM-Refine pipeline (ablation) | SC-wLS (dlt+ref) · 7Scenes Stairs | median translation error (m): 0.067 -> 0.058 | Improved |
| add end-to-end stage to LM-Refine pipeline (ablation) | SC-wLS (dlt+ref) · 7Scenes Pumpkin | median translation error (m): 0.045 -> 0.051 | Hurt |
| replace DLT-initialized LM-Refine with DSAC* post-processing (ablation) | SC-wLS (dlt+e2e) · 7Scenes average recall | recall (%): 64.7 -> 80.0 | Improved |
| replace DSAC* with SC-wLS weighted DLT feed-forward | MS-Transformer · 7Scenes Chess | median translation error (m): 0.11 -> 0.029 | Improved |
| add LM-Refine to SC-wLS vs DSAC* | DSAC* w/o model · Old Hospital | median translation error (m): 0.21 -> 0.11 | Improved |
Closest papers by meaning
- Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization (CVPR 2025 · similarity 0.87)
- GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024 · similarity 0.87)
- Map-Relative Pose Regression for Visual Re-Localization (CVPR 2024 · similarity 0.87)
- ConDo: Continual Domain Expansion for Absolute Pose Regression (AAAI 2025 · similarity 0.86)
- R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization (CVPR 2025 · similarity 0.86)
- HyperPose: Hypernetwork-Infused Camera Pose Localization and an Extended Cambridge Landmarks Dataset (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.
