Skip to content
/ GTM Public
forked from Ultimecia1463/GTM

Real-time hand gesture detection using a model from Google's Teachable Machine and TensorFlow.js, allowing gesture-based control of on-screen elements via webcam input.

Notifications You must be signed in to change notification settings

omendrag/GTM

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Gesture Detection Using Teachable Machine

This project implements gesture detection using a model trained with Google's Teachable Machine. The application detects hand gestures (such as fist, palm, and directional movements) via webcam input and allows interaction with an on-screen element (a red square), which moves based on the detected gestures.

Features

  • Real-time gesture detection using a pre-trained model from Teachable Machine.
  • Webcam-based interface to detect hand gestures.
  • Moves an on-screen red square based on detected gestures:
    • Fist: Move right.
    • Palm: Move left.
    • Thumb up: Move up.
    • Thumb down: Move down.
  • Simple and intuitive UI with webcam feed display.

Technologies Used

  • Teachable Machine by Google
  • TensorFlow.js for pose estimation and gesture recognition
  • HTML5 Canvas for webcam feed and rendering keypoints/skeleton
  • JavaScript for real-time webcam control and DOM manipulation

How It Works

  1. The model is loaded from Teachable Machine and integrated using TensorFlow.js.
  2. The webcam is initialized, and the feed is displayed inside a canvas element.
  3. Hand gestures are detected in real-time, and the position of a red square on the screen is updated based on the detected gestures.

Installation

  1. Clone the repository:
    git clone https://github.com/your-username/gesture-detection.git
  2. Open index.html in your preferred browser.

No additional installation steps are needed, as the project uses a CDN for TensorFlow.js and Teachable Machine.

Live Demo

View the live demo here

Usage

  • Open the app.
  • Ensure you allow access to the webcam when prompted.
  • Use hand gestures (fist, palm, thumb up, thumb down) to move the red square on the screen.

Future Enhancements

  • Add more gestures for additional functionality.
  • Improve gesture recognition accuracy.
  • Add sound or other feedback mechanisms for detected gestures.

About

Real-time hand gesture detection using a model from Google's Teachable Machine and TensorFlow.js, allowing gesture-based control of on-screen elements via webcam input.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 51.1%
  • CSS 29.6%
  • HTML 19.3%