Purpose: to train a computer to recognize the number represented by human’s hand gestures

Set up an image dataset which contains gestures performed by 4 different people, each performing 11 different gestures repeated 30 times each, for a total of 1320 samples. The raw images are depth map images of the original ones.

Use three models to train the dataset, which are KNN, SVM, and CNN. And compare the performance of these three methods.