S010MON/cross-view-localisation-WITW

Using ML to match ground level photos to overhead imagery.

Where in the World

The Where in the World (WITW) project has developed tools and capabilities for cross-view image geolocalization (CVIG). CVIG is the process of matching a ground-level outdoor photograph to its location using overhead imagery of potential locations as the only point of reference.

In this repo, the model folder contains the deep learning model and instructions on how to run it, while the tools folder contains various software tools used for the project.

More information on Where in the World can be found in our blog post series:

• Where in the World, Part 1: A New Dataset for Cross-view Image Geolocalization
• Where in the World, Part 2: A Deep Learning Model for Cross-view Image Geolocalization
• Where in the World, Part 3: Further Analysis of Cross-view Image Geolocalization