Hand (or shape) tracking implementation of Kris Kitani's per-pixel segmentation algorithm, for OpenFrameworks
README
OpenFrameworks based machine vision application for tracking a human hand on a contrasting background, and to train a Support Vector Machine (SVM) to classify static hand poses.
This application requires the following addons:
- ofxConvexHull
- ofxCv
- ofxGrt
- ofxGui
- ofx.JSON
- ofxOpenCv
- ofxOSC
- ofxPerPixelSegment
Steps for getting this project running:
- install OpenFrameworks 0.93
- install all the addons listed above
- Clone this repository into .../of/apps/myApps/
- Open 'ofxRGBHandTracK.xcodeproj'
- Use the OFXCodeMenu to add all the addons to a the project
- Build (make sure to build 'ofxRGBHandTracK' not 'openFrameworks')
- use the openFrameworks Project builder to make a new project; include all the necessary addons
- replace the three files inside the generated projects .../src folder with the ones from this repository
ofxAddons