Low cost hand gesture recognition software using MATLAB
Final year Electrical and Information Engineering students Nabeel Vandayar and Timothy McBride from the University of Witwatersrand discuss their project which provided solutions to everyday life challenges.
“The core component of the entire system is the machine learning model or the neural network, we did experiments on a few other platforms from Python to using Tensor Flow and Pi Torch but the start-up time or learning curve on those platforms were high and the support was lacking. The main toolbox chosen was the Neural Networks toolbox in MATLAB because the support was immense, and the existing examples enabled us to get 50% of the way without much work except for modifying the models slightly and changing a few lines of code.”
