This is a project I undertook while doing my AI masters at the University of North Texas. A pretrained model that predicts 3D key points from an image of a hand for hand tracking. This model is used to interact with a virtual UI.
(Open
Hand) Neutral.
(Closed
Hand) To select the object.
(Lamma) To
trigger speaking the sentence.
(Shaka) To
clear the sentence.
(Spiderman) To go back in the UI.