A simple WebSocket-based video streaming system from Raspberry Pi to a web interface.

• Backend: FastAPI WebSocket server that receives video from Pi and broadcasts to viewers
• Pi Connection: Python client that captures video from Pi Camera and streams via WebSocket
• Frontend: Simple HTML/JS viewer embedded in the backend

1. Install dependencies:

cd backend
pip install -r requirements.txt

2. Run the server:

python main.py

The server will start on http://0.0.0.0:8000

3. Open a browser and go to http://localhost:8000 to view the stream

1. Install system dependencies (if not already installed):

sudo apt-get update
sudo apt-get install -y python3-picamera2 python3-libcamera python3-pip libcap-dev

2. Create virtual environment with system-site-packages (required for libcamera access):

python3 -m venv --system-site-packages .venv
source .venv/bin/activate

3. Install Python dependencies:

pip install -r requirements.txt

3. Set up SSH tunnel from Pi to your desktop:

# On the Raspberry Pi, create SSH tunnel to desktop
ssh -L 8000:localhost:8000 your_username@your_desktop_ip

Keep this SSH connection open in a separate terminal.

4. Enable the camera:

sudo raspi-config
# Navigate to: Interface Options -> Camera -> Enable

5. Run the stream client (in a new terminal on the Pi):

python3 stream.py

The client will connect to localhost:8000 which tunnels to your desktop.

1. Pi Side:

   • Captures video frames from the Pi Camera
   • Encodes frames as JPEG
   • Sends frames via WebSocket to /ws/pi endpoint

2. Backend Side:

   • Receives frames from Pi at /ws/pi
   • Broadcasts frames to all connected viewers at /ws/viewer
   • Serves a simple HTML viewer at the root URL

3. Viewer Side:

   • Connects to /ws/viewer via WebSocket
   • Receives JPEG frames as binary data
   • Displays frames as images in the browser

• GET / - HTML viewer interface
• WS /ws/pi - WebSocket endpoint for Pi to send stream
• WS /ws/viewer - WebSocket endpoint for viewers to receive stream
• GET /health - Health check endpoint

• Default port: 8000
• Change in the uvicorn.run() call

• SERVER_HOST: localhost (via SSH tunnel)
• WIDTH/HEIGHT: Video resolution (default: 640x480)
• FPS: Frame rate (default: 20 fps)

The Pi connects to your desktop via SSH tunnel:

# From Pi to Desktop
ssh -L 8000:localhost:8000 user@desktop_ip

This forwards port 8000 on the Pi to port 8000 on your desktop.

• Ensure camera is enabled: sudo raspi-config
• Check camera connection: libcamera-hello
• Verify picamera2 installation: python3 -c "import picamera2"

• Ensure SSH tunnel is active and running
• Check that backend server is running on desktop
• Verify SSH connection: ssh user@desktop_ip
• Test tunnel: curl http://localhost:8000/health (from Pi)

• Lower FPS in stream.py
• Reduce resolution (e.g., 320x240)
• Check network bandwidth

• Add audio streaming
• Multiple camera support
• Recording functionality
• Authentication
• Better error handling
• Stream quality controls