Skip to content

Latest commit

 

History

History
19 lines (10 loc) · 1.42 KB

README.md

File metadata and controls

19 lines (10 loc) · 1.42 KB

Developing a Computer Vision Learning Algorithm for the Game Geo Guesser

Geo Guesser is a popular online game that challenges players to guess the location of a randomly selected place in the world based on a Google Street View image.

The objective of this project is to develop a computer vision algorithm trained to recognize and predict the location of a place based on its visual features, such as architecture, vegetation, or natural landmarks.

The algorithm to be developed, will utilize deep learning techniques, specifically Convolutional Neural Networks (CNNs).

The scope of the project will determine the method used for training the algorithm, which could be through conventional deep learning with a substantial dataset of Google Street View images and geographic locations, or through reinforcement learning techniques by letting the algorithm play the game.

A brief survey of available resources presents a wide range of potential training datasets, including options like (https://www.kaggle.com/datasets/ubitquitin/geolocation-geoguessr-images-50k).

In conclusion, it is important to acknowledge that a project of this scope and complexity will require a significant amount of time and resources. The desired outcome may present challenges, but with a well-structured plan, it is feasible.

Done by Hussein Abdulreda, Amir Leidel, Daniel Richter