Research map / 3D vision / Depth and stereo
SANet: Scene Agnostic Network for Camera Localization
Luwei Yang, Ziqian Bai, Chengzhou Tang, Honghua Li, Yasutaka Furukawa, Ping Tan
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 |
|---|---|---|---|
| use the learned QSR and Fusing network instead of explicit angular-similarity feature matching (ablation) | SANet · 7Scenes | poses within (5 deg, 5cm): 63.8% -> 68.2% | Improved |
| compare SANet to InLoc on every 10th test frame | SANet · 7Scenes | poses within (5 deg, 5cm): 66.3% -> 67.9% | Improved |
| use scene-agnostic SANet instead of scene-specific DSAC++ | SANet · 7Scenes | poses within (5 deg, 5cm): 76.1% -> 68.2% | Hurt |
| use SANet instead of InLoc pose estimation | SANet · 7Scenes | total time per query (s): 9.55s -> 0.54s | Improved |
| use SANet instead of Active Search on Stairs | SANet · 7Scenes Stairs | median translation error (m): 0.09m -> 0.16m | Hurt |
| use SANet instead of InLoc on Great Court | SANet · Cambridge Great Court | median translation error (m): 1.20m -> 3.28m | Hurt |
| use SANet instead of InLoc on King's College | SANet · Cambridge King's College | median translation error (m): 0.46m -> 0.32m | Improved |
Later papers that compared with this method
1 later accepted paper reports SANet as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- Adaptive Density Map Generation for Crowd Counting (ICCV 2019 · better on 1 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- InLoc — InLoc: Indoor Visual Localization With Dense Matching and View Synthesis (CVPR 2018) · better on 1 of 8 shared results
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
- Learning Camera Localization via Dense Scene Matching (CVPR 2021 · similarity 0.91)
- Learning Less Is More - 6D Camera Localization via 3D Surface Regression (CVPR 2018 · similarity 0.88)
- Hierarchical Scene Coordinate Classification and Regression for Visual Localization (CVPR 2020 · similarity 0.87)
- Geometry-Aware Learning of Maps for Camera Localization (CVPR 2018 · similarity 0.86)
- Point-SLAM: Dense Neural Point Cloud-based SLAM (ICCV 2023 · similarity 0.86)
- Loopy-SLAM: Dense Neural SLAM with Loop Closures (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.
