Research map / 3D vision / Depth and stereo
ConGeo: Robust Cross-view Geo-localization across Ground View Variations
Li Mi, Chang Xu, Javiera Castillo Navarro, SYRIELLE MONTARIOL, Wen Yang, Antoine Bosselut, Devis Tuia
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 |
|---|---|---|---|
| train one model with ConGeo contrastive objectives instead of North-aligned Sample4Geo | Sample4Geo · CVUSA (FoV=360) | R@1: 16.3 -> 85.2 | Improved |
| add cross-view contrastive loss on top of single-view losses (ablation) | ConGeo · CVUSA (FoV=90) | R@1: 20.5 -> 55.9 | Improved |
| add aerial-view single contrastive loss (ablation) | ConGeo · CVUSA (FoV=180) | R@1: 4.1 -> 15.7 | Improved |
| add ground-view shift contrastive loss (ablation) | ConGeo · CVUSA (FoV=180) | R@1: 15.7 -> 44.1 | Improved |
| add FoV cropping as an additional ground-view single contrastive target (ablation) | ConGeo · CVUSA (FoV=180) | R@1: 44.1 -> 37.9 (FoV=180); 17.8 -> 20.5 (FoV=90) | Mixed |
| replace ConGeo contrastive objective with targeted data augmentation (shift, FoV, rotate) (ablation) | Sample4Geo · CVUSA (North-aligned) | R@1: 98.7 -> 88.9 to 94.4 | Hurt |
| ConGeo versus data augmentation on unseen ground view variations (ablation) | Sample4Geo · CVUSA unseen variations (average) | average R@1: 36.8 -> 57.8 | Improved |
| plug ConGeo into TransGeo versus data augmentation baseline | TransGeo · CVUSA (FoV=360) | R@1: TransGeo+DA 75.9 vs ConGeo 52.7 at 360; 18.7 vs 26.9 at 90 | Mixed |
| ConGeo versus Sample4Geo North-aligned test | Sample4Geo · CVUSA North-aligned | R@1: 98.7 -> 98.3 | Mixed |
| apply ConGeo on VIGOR cross-area | Sample4Geo · VIGOR cross-area (FoV=360) | R@1: 9.0 -> 16.2 | Improved |
Later papers that compared with this method
3 later accepted papers report ConGeo as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- SinGeo: Unlock Single Model's Potential for Robust Cross-View Geo-Localization (CVPR 2026 · better on 9 of 10 shared results)
- Sector-Level Cross-View Geo-Localization with Implicit Orientation via Azimuthal Scanning (ECCV 2026 · better on 10 of 10 shared results)
- First Learn, Then Review: Human-Like Continual Learning for Cross-View Geo-Localization with Limited Field of View (AAAI 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Sample4Geo — Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisation (ICCV 2023) · better on 5 of 5 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
- SinGeo: Unlock Single Model's Potential for Robust Cross-View Geo-Localization (CVPR 2026 · similarity 0.92)
- Cross-View Geo-Localization via Learning Disentangled Geometric Layout Correspondence (AAAI 2023 · similarity 0.91)
- Geo2: Geometry-Guided Cross-view Geo-Localization and Image Synthesis (CVPR 2026 · similarity 0.90)
- From Coarse to Fine: A Matching and Alignment Framework for Unsupervised Cross-View Geo-Localization (AAAI 2025 · similarity 0.90)
- Aligning Geometric Spatial Layout in Cross-View Geo-Localization via Feature Recombination (AAAI 2024 · similarity 0.90)
- Semantic Ambiguity Modeling and Propagation for Fine-Grained Visual Cross View Geo-Localization (AAAI 2025 · similarity 0.88)
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.
