Research map / 3D vision / Depth and stereo
Semantic Match Consistency for Long-Term Visual Localization
Carl Toft, Erik Stenborg, Lars Hammarstrand, Lucas Brynte, Marc Pollefeys, Torsten Sattler, Fredrik Kahl
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 semantic-consistency-weighted RANSAC sampling (ablation) | P3P RANSAC · CMU Seasons Urban | % queries within (0.25m, 2deg): 42.5 -> 75.2 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P3P RANSAC · CMU Seasons Suburban | % queries within (0.25m, 2deg): 11.5 -> 44.6 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P3P RANSAC · CMU Seasons Park | % queries within (0.25m, 2deg): 9.3 -> 30.4 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P2P RANSAC · CMU Seasons Urban | % queries within (0.25m, 2deg): 72.9 -> 81.7 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P2P RANSAC · CMU Seasons Suburban | % queries within (0.25m, 2deg): 41.3 -> 55.4 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P2P RANSAC · CMU Seasons Park | % queries within (0.25m, 2deg): 29.7 -> 39.5 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P3P RANSAC · RobotCar Seasons all day | % queries within (0.25m, 2deg): 47.0 -> 50.6 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P2P RANSAC · RobotCar Seasons all day | % queries within (0.25m, 2deg): 34.1 -> 35.4 | No clear effect |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P3P RANSAC · RobotCar Seasons all night | % queries within (0.25m, 2deg): 0.5 -> 7.6 | Improved |
| add semantic-consistency-weighted RANSAC sampling (ablation) | P2P RANSAC · RobotCar Seasons all night | % queries within (0.25m, 2deg): 5.1 -> 13.0 | Improved |
| replace the P3P solver with the gravity-aided P2P solver (ablation) | unweighted RANSAC · CMU Seasons Urban | % queries within (0.25m, 2deg): 42.5 -> 72.9 | Improved |
| replace the P3P solver with the gravity-aided P2P solver (ablation) | semantic-weighted RANSAC · RobotCar Seasons all day | % queries within (0.25m, 2deg): 50.6 -> 35.4 | Hurt |
Closest papers by meaning
- Semantic Visual Localization (CVPR 2018 · similarity 0.90)
- Weatherproofing Retrieval for Localization with Generative AI and Geometric Consistency (ICLR 2024 · similarity 0.89)
- Change-Robust Online Topological Memory for Long-Term Relocalization and Semantic Navigation (NeurIPS 2026 · similarity 0.88)
- Semantic Ambiguity Modeling and Propagation for Fine-Grained Visual Cross View Geo-Localization (AAAI 2025 · similarity 0.87)
- RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations (CVPR 2025 · similarity 0.87)
- On the Estimation of Image-matching Uncertainty in Visual Place Recognition (CVPR 2024 · similarity 0.86)
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.
