/hand-gesture-recognition

Hand tracking and gesture recognition using neural networks and traditional model pipeline

Primary LanguagePythonMIT LicenseMIT

BSc graduation thesis

Hand tracking and gesture recognition using neural networks and traditional model pipeline

This paper presents a solution for hand tracking and gesture recognition using a pipeline consisting of the Mediapipe Hands model and a classification model (neural network, SVM, and random forest). The system was tested on the German Sign Language dataset.

Methodology

Tracker Demo

Demo

Hand Tracking (MediaPipe Hands)

Tracker Demo

Hand Tracking + NN classification

Tracker Demo

Hand Tracking + SVM classification

Tracker Demo

Hand Tracking + RF classification

Tracker Demo