Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Visual Localization Interoperability: Cross-Feature for Collaborative Visual Localization and Mapping
Alberto Jaenal, Paula Carbó Cubero, José Araújo, André Mateus
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 Cross-Feature patch-based training instead of Cross-Descriptor | SIFT map, SuperPoint query · Aachen Day (0.25m, 2deg) | % localized queries: 20.4 -> 56.3 | Improved |
| Use Cross-Feature instead of Cross-Descriptor with a different detector (OrFAST:BRIEF) | SuperPoint map, ORB query · Aachen Day (0.25m, 2deg) | % localized queries: 1.0 -> 13.5 | Improved |
| Use Cross-Feature instead of Cross-Descriptor on Cambridge | SIFT map, SuperPoint query · Cambridge GreatCourt | Median position error (m): 0.96 -> 0.43 | Improved |
| Use Cross-Feature instead of Cross-Descriptor on 7Scenes | SIFT map, SuperPoint query · 7Scenes Stairs | Median position error (cm): 23.98 -> 20.62 | Improved |
| Use Cross-Feature instead of image reconstruction | SIFT map, SuperPoint query · Aachen Day (0.25m, 2deg) | % localized queries: 34.0 -> 56.3 | Improved |
| Use Cross-Feature instead of image reconstruction | SuperPoint map, SIFT query · Aachen Day (0.25m, 2deg) | % localized queries: 77.8 -> 43.1 | Hurt |
| Use Cross-Feature instead of geometry-only GoMatch | SuperPoint map, SIFT query · Aachen Day (0.25m, 2deg) | % localized queries: 0.2 -> 43.1 | Improved |
| Use heterogeneous query features instead of homogeneous SIFT | CFemb, SIFT map · Aachen Day (0.25m, 2deg) | % localized queries: 80.8 -> 56.3 | Hurt |
| Change query feature from SuperPoint to DISK | CFemb, SIFT map · 7Scenes Stairs | Median position error (cm): 20.62 -> 48.19 | Hurt |
| Change query from ORB to SuperPoint on SuperPoint-map Aachen night | CFemb · Aachen Night (0.25m, 2deg) | % localized queries: 17.3 -> 6.1 | Hurt |
| Use collaborative merged map instead of SIFT-only map | CFemb, query SuperPoint · Cambridge GreatCourt | Median position error (m): 0.43 -> 0.30 | Improved |
| Use Cross-Feature on collaborative map instead of Cross-Descriptor | query SuperPoint · Cambridge GreatCourt (collaborative) | Median position error (m): 0.85 -> 0.30 | Improved |
Closest papers by meaning
- Cross-Descriptor Visual Localization and Mapping (ICCV 2021 · similarity 0.93)
- Aligning Geometric Spatial Layout in Cross-View Geo-Localization via Feature Recombination (AAAI 2024 · similarity 0.90)
- From Coarse to Fine: A Matching and Alignment Framework for Unsupervised Cross-View Geo-Localization (AAAI 2025 · similarity 0.89)
- Enhancing Visual Localization with Cross-Domain Image Generation (ICML 2025 · similarity 0.89)
- CroCoDL: Cross-device Collaborative Dataset for Localization (CVPR 2025 · similarity 0.88)
- 360Loc: A Dataset and Benchmark for Omnidirectional Visual Localization with Cross-device Queries (CVPR 2024 · similarity 0.87)
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.
