Research map / Vision-language and multimodal / Human and embodied modeling
Rethinking Visual Geo-Localization for Large-Scale Applications
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 non-adjacent spatial and orientation-based CosPlace groups (ablation) | CosPlace · SF-XL validation | Recall@1: 77.1 -> 90.9 | Helped |
| Add orientation labels and orientation-based class separation (ablation) | CosPlace · SF-XL validation | Recall@1: 85.8 -> 90.9 | Helped |
| Add orientation group separation (L=2) (ablation) | CosPlace · SF-XL validation | Recall@1: 88.2 -> 90.9 | Helped |
| Add spatial group separation (N=5) (ablation) | CosPlace · SF-XL validation | Recall@1: 77.1 -> 90.9 | Helped |
| Train CosPlace with fewer images (ablation) | CosPlace · Pitts250k | Recall@1: 90.4 -> 86.0 -> 82.1 -> 73.3 | Hurt |
| Train using CosPlace classification instead of contrastive retrieval learning | Visual geo-localization · SF-XL test v1 | R@1: 51.2 -> 64.7 | Helped |
| Use a 512-D CosPlace descriptor instead of a 4096-D descriptor | Visual geo-localization · SF-XL test v1 | R@1: 51.2 -> 64.7 | Helped |
Closest papers by meaning
- Recognition through Reasoning: Reinforcing Image Geo-localization with Large Vision-Language Models (NeurIPS 2025 · similarity 0.88)
- RELOCATE: A Simple Training-Free Baseline for Visual Query Localization Using Region-Based Representations (CVPR 2025 · similarity 0.87)
- OpenStreetView-5M: The Many Roads to Global Visual Geolocation (CVPR 2024 · similarity 0.87)
- R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization (CVPR 2025 · similarity 0.87)
- GRE Suite: Geo-localization Inference via Fine-Tuned Vision-Language Models and Enhanced Reasoning Chains (NeurIPS 2025 · similarity 0.87)
- GeoReasoner: Geo-localization with Reasoning in Street Views using a Large Vision-Language Model (ICML 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.