AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Cross-View Image Matching for Geo-Localization in Urban Environments

Yicong Tian, Chen Chen, Mubarak Shah

CVPR 2017 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 228 citations (OpenAlex)

Check 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 changedWhereMeasuredEffect
Fine-tune the Siamese network on annotated cross-view building pairs (ablation)ImageNet-pretrained AlexNet Siamese network · Cross-view building matching test pairsaverage precision (AP): 0.11 -> 0.32Improved
Use four directional images at one location instead of one query image (ablation)Building-matching geo-localization · Cross-view geo-localizationgeo-localization accuracyImproved
Match detected buildings rather than matching full images (ablation)Cross-view geo-localization · Cross-view geo-localizationgeo-localization accuracyImproved
Use dominant sets instead of GMCP for multiple-neighbor matching (ablation)Cross-view geo-localization · Bird's-eye-view query, k=10geo-localization accuracyImproved
Use dominant sets instead of GMCP for multiple-neighbor matching (ablation)Cross-view geo-localization · Multiple-neighbor matchingruntimeImproved
Use deep building matching instead of SIFT matching (ablation)Cross-view geo-localization · Cross-view geo-localizationgeo-localization accuracyImproved
Use building matching instead of random reference-image selection (ablation)Cross-view geo-localization · Cross-view geo-localizationgeo-localization accuracyImproved
Increase the number of nearest reference neighbors considered for street-view queries (ablation)Multiple-neighbor matching · Street-view query, bird's-eye-view referencegeo-localization accuracyMixed
Train and test the method across cities, with Pittsburgh and Orlando for training and Manhattan for testingProposed cross-view geo-localization method · Manhattangeo-localization accuracyImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy