This web application utilizes MediaPipe Pose and ml5.js neural network to recognize and classify yoga poses in real-time through a webcam. The program displays the recognized pose along with an accuracy score.

• A web browser with webcam access.
• A stable internet connection.

• Clone the repository to your local machine:

   git clone https://github.com/your-username/yoga-pose-recognition.git
  

• Open the index.html file in your preferred web browser.

• Ensure your webcam is enabled and accessible.

• Strike a yoga pose in front of the webcam, and the application will display the recognized pose and accuracy score.

The application recognizes yoga poses in real-time using the MediaPipe Pose library. Detected poses are then processed and classified using a pre-trained neural network model.

The program uses ml5.js to load and utilize a pre-trained neural network model (model/model.json). The model has been trained to classify yoga poses based on input features.

If you find any issues or have suggestions for improvements, feel free to open an issue or submit a pull request.