Skip to content

ESE 650: Learning in Robotics, Project 3, Gesture Recognition using Hidden Markov Models and IMU Data

License

Notifications You must be signed in to change notification settings

NitinJSanket/ESE650Project3

Repository files navigation

Gesture Recognition using Hidden Markov Models on IMU Data

Problem Statement

Use Hidden Markov Model on IMU data (accelerometer and gyroscope) to classify gestures.

Usage Guide:

  1. Run WrapperTrainandTest.m for running both training and testing.
  2. Run WrapperTestOnly.m for testing only (Change TestPath as needed) 3.You'll need a folder called Outputs to save the plots or else the code will throw an error.

Report:

You can find the report here.

References:

  1. Rabiner, Lawrence R. "A tutorial on hidden Markov models and selected applications in speech recognition." Proceedings of the IEEE 77.2 (1989): 257-286.
  2. http://www.mathworks.com/help/signal/examples/signal-smoothing.html?prodcode=SG&language=en#zmw57dd0e318

About

ESE 650: Learning in Robotics, Project 3, Gesture Recognition using Hidden Markov Models and IMU Data

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages