Embedding AI into Android Games

March 11, 2019 1:00 pm GST

In recent years, one of the biggest applications of machine learning has been in image classification — the ability for a computer to intelligently recognize an object. Such work is being done at the greatest technology companies, and it’s since spawned this unwarranted notion of exclusivity. It is also widely used in gaming industry nowadays. We’ve crafted a narrative that in order to get your hands dirty with machine learning and artificial intelligence in android games. In this workshop, you’ll learn how to build a fully-functioning image classifier that can recognize objects with insanely high probabilities in less than 45 minutes. See the description of the game below:

AI Treasure Hunt Game

An Android Game that uses IBM Watson Visual Recognition service. This is a Treasure Hunt Game, in which the player will be given hints to find a place in Dubai (famous landmarks etc.) and take a photo of it to pass a level. To determine if the player passes the level, a custom visual recognition model is built with a dataset of images. This model classifies the photo captured by the player. If the image captured is of the correct place, next level starts. The models are trained with Watson Studio. You can also add security to or track players of the game by using App ID service in IBM Cloud.

in5 Tech, King Salman Bin Abdulaziz Al Saud Street, near Tram Station and Cordoba Residence, Dubai, United Arab Emirates

Legend