Automatic Generation of Pictorial Maps from Photos on Social Media to Represent Regional Features
Wentong Zhou, Taketoshi Ushiama · 2020
In recent years, with the spread of smartphones and SNSs, billions of photos have been posted on SNSs. Photographs from different locations show different charms of each region. This paper introduces a system for automatic generation of pictorial maps using geotagged photographs posted on Flickr. The proposed system can extract the most representative photograph of each region automatically and can generate pictorial maps with the extracted photographs. Generated pictorial maps can be used to help people to discover the unique charms and cultures of a region in an easy-to-understand manner and the proposed system can also be applied to various web services such as geographic search engines, travel apps and so on.