Development is in working progress
This code is no more developped. Check out the new ofxSequence.
Hidden Markov Models is a powerful statistical tool for model time-varying pattern. Gestures recognition is a good application for this kind of machine learning automatic pattern recognition. It consists in two phases: supervised learning with examples (training) and recognition (classification).
This C++ project provide a simple graphical user interface for boh training and classification.
It is a projected developped with:
- OPEN FRAMEWORKS, thanks to the great community
- ofxhGUI, my fork of an event-based graphical interface for OF devlopped by Bernard Geyer
- [Armadillo](C++ linear algebra library), lightweight and fast library for matrix calculi
Hidden Markov Model code is based on HHM toolbox from Kevin Murph
For now it is been tested only under Ubuntu 10.10.