An example of how to integrate the Gesture Recognition Toolkit into an iPhone app

The Gesture Recognition Toolkit is a "cross-platform, open-source, C++ machine learning library designed for real-time gesture recognition". This repository contains the project outlined in my blog posts, Integrating the GRT into an iPhone app and Machine-Learning powered Gesture Recognition on iOS.

Here are some of the gestures I was able to train the system to recognize:

SwiftR is used for visualizing the acellerometer data.

Since I worked on this project, there's been a lot of advancemenrts in Apple's CoreML framework, including the ability to create Motion Activity Classifiers using CreateML. If you're looking to create a gestural recognition system for iOS from scratch, I would recommend taking a look at the linked Apple talk.

Additional resources:

• Activity Classification with Create ML, CoreML3, and Skafos: Part 1

• GestureAI-CoreML-iOS