Skip to content

Latest commit

 

History

History
31 lines (20 loc) · 1.79 KB

README.md

File metadata and controls

31 lines (20 loc) · 1.79 KB
Raw

Hand Gesture Detection using Machine Learning

About

This repository contains the winning solution developed for the SIH (Smart India Hackathon) Hackathon, focusing on the topic of hand gesture detection using machine learning. The solution utilizes cutting-edge technologies, including Mediapipe, deep learning techniques such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), and a Google Chrome extension for real-time gesture visualization during virtual meetings.

Features

  • Accurate hand gesture detection powered by CNNs and RNNs.
  • Seamless integration with Mediapipe for precise gesture analysis.
  • Real-time interpretation of both spatial and temporal aspects of hand movements.
  • Integration as a Google Chrome extension, enhancing virtual meetings with interactive gestures.
  • Transforms conventional video conferencing into engaging and dynamic discussions.

Technologies Used

  • Mediapipe: Framework for real-time multimedia processing.
  • Deep Learning: Utilizes Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for gesture interpretation.
  • Google Chrome Extension: Overlaying real-time gesture visualizations onto virtual meetings.

Getting Started

  1. Clone the repository.
  2. Install the required dependencies listed in requirements.txt.
  3. Explore the codebase to understand the implementation of gesture detection and integration with Mediapipe.
  4. Load the provided pre-trained models for immediate usage or train your own models using the provided datasets.
  5. Integrate the solution into Google Chrome as an extension following the instructions in the extension/ directory.

Feel free to contribute to the project by opening issues or pull requests. Your feedback and contributions are highly appreciated!