Virtual Canvas

Overview

Virtual Canvas is an interactive drawing application that allows users to draw using hand gestures. Utilizing MediaPipe for hand tracking, this project enables gesture-based drawing, erasing, and shape selection using a webcam. The application provides an intuitive and touch-free way to create digital sketches.

Features

• Hand Tracking: Uses MediaPipe Hands to detect and track hand movements.
• Gesture-Based Drawing: Supports freehand drawing, shape selection, and erasing.
• Multiple Colors & Shapes: Users can switch between different colors and shapes such as circles, rectangles, and ellipses.
• Dynamic Brush & Eraser Thickness: Adjusts based on finger distance.
• Real-time Processing: Smooth interaction and minimal latency.

Technologies Used

• Python
• OpenCV
• MediaPipe
• NumPy
• os (for managing images and assets)

Installation

1. Clone the Repository

   git clone https://github.com/SagarMaddela/Virtual-Canvas.git
   cd Virtual-Canvas

2. Install Dependencies Ensure you have Python installed (preferably 3.7+), then install the required packages:

   pip install opencv-python mediapipe numpy

Usage

1. Run the Application

   python Main.py

2. Interact Using Gestures
   • Index & Middle Finger Up → Selection Mode (Choose colors/shapes/eraser)
   • Only Index Finger Up → Drawing Mode (Draw on the canvas)
   • Index & Thumb Close Together → Adjust Brush/Eraser Size

Project Structure

VirtualCanvas/
│-- MY Header/          # Folder containing UI overlay images
│-- Main.py             # Main script to run the Virtual Canvas
│-- HandTrackingModule.py  # Module for hand tracking functionalities
│-- README.md           # Project documentation

Troubleshooting

• Camera Not Opening?

  • Ensure your webcam is connected and functional.
  • Try changing video = cv2.VideoCapture(1) to cv2.VideoCapture(0).

• Hand Not Detected?

  • Make sure your hand is well-lit and visible to the camera.
  • Adjust detectionCon in handDetector() for better recognition.

Contributing

Feel free to fork this repository and submit pull requests for improvements.

License

This project is open-source and available under the Apache License.

Author

Developed by Sagar Maddela

For any queries, feel free to reach out on GitHub.

---

Happy Drawing! 🎨