In the popular online game Geoguessr, the player is shown a random image from Google Street View and is tasked with guessing their location on the globe as accurately as possible. In this project, we seek to solve a simplified version of this problem but using a strategy often used by professional Geoguesser players: using man-made features (for example, traffic lights) to accurately guess a city.

We use the publicly available GSV-Cities Dataset, which consists of around 500k street-view images taken in 23 different cities. We then use CNN trained on the images and features extracted from the images to make our mode. The backbone of this CNN is a pre-trained model named MobileNetV2.