Skip to content

Latest commit

 

History

History
65 lines (42 loc) · 1.76 KB

README.md

File metadata and controls

65 lines (42 loc) · 1.76 KB
Raw

Hand Sign Detection using TensorFlow Lite and Flask

This project demonstrates hand sign detection using TensorFlow Lite and Flask. It captures video frames, performs hand sign detection using a pre-trained TensorFlow Lite model, adds a unique number to the frame, and displays the processed frames in a web interface using Flask.

Features

  • Capture video frames and perform hand sign detection.
  • Display processed frames in a web interface using Flask.
  • Add unique identifiers to each processed frame.
  • Store processed frames on the server.

Getting Started

Prerequisites

  • Python 3.7 or higher
  • Virtualenv (optional but recommended)

Installation

  1. Clone the repository:

    git clone https://github.com/your-username/hand-sign-detection.git
cd hand-sign-detection

  2. (Optional) Create and activate a virtual environment:

    python3 -m venv venv
source venv/bin/activate

  3. Install the required dependencies from requirements.txt:

    Copy code
pip install -r requirements.txt

Running the Application

  1. Start the Flask application:
python app.py

  1. Open your web browser and navigate to http://127.0.0.1:5000 to access the web interface.

  2. Follow the instructions on the web interface to capture and process hand sign frames.

Usage

  • Follow the on-screen instructions to capture video frames.
  • The processed frames will be displayed with unique numbers added.
  • You can view the stored processed frames in the processed_frames folder.

Contributing

Contributions are welcome! If you'd like to contribute to this project, please follow the guidelines in CONTRIBUTING.md.

License

This project is licensed under the MIT License - see the LICENSE file for details.