Parkour Spot ID: Feature Matching in Satellite and Street view images using Deep Learning

TL;DR

This paper presents a deep learning framework that matches satellite and street view images to identify previously unindexed locations, demonstrated by discovering new Parkour spots on a university campus.

Contribution

It introduces a novel approach combining satellite and street view imagery for location classification, enabling discovery of unindexed places using deep learning.

Findings

Identified over 25 new Parkour spots.

Achieved true positive rate above 60%.

Effective in locating unindexed locations.

Abstract

How to find places that are not indexed by Google Maps? We propose an intuitive method and framework to locate places based on their distinctive spatial features. The method uses satellite and street view images in machine vision approaches to classify locations. If we can classify locations, we just need to repeat for non-overlapping locations in our area of interest. We assess the proposed system in finding Parkour spots in the campus of Arizona State University. The results are very satisfactory, having found more than 25 new Parkour spots, with a rate of true positives above 60%.