camera_routes.md

Camera Routes Example

This example demonstrates video processing pipelines using Taskinity DSL, including frame capture, object detection, and video streaming capabilities.

Prerequisites

• Python 3.8 or higher
• OpenCV (pip install opencv-python)
• Taskinity DSL installed
• Docker (for containerized processing)
• Webcam or video file for input

Example Files

• camera_routes.yaml - Main configuration file
• docker-compose.yml - Processing services
• models/ - Pre-trained models
• scripts/ - Processing scripts

Running the Example

Using Make (Recommended)

# Start the camera processing pipeline
make run-example EXAMPLE=camera

# View the processed video stream
make view-camera

# Stop the example
make stop-example EXAMPLE=camera

Manual Execution

1. Start the processing services:

   docker-compose -f examples/docker-compose.yml up -d

2. Start the router:

   python -m src.camel_router.cli --config examples/camera_routes.yaml

Example Configuration

routes:
  # Capture frames from camera
  - from:
      uri: "direct:camera"
    steps:
      - process:
          ref: "frameCapture"
      - to: "direct:objectDetection"

  # Process frames with object detection
  - from:
      uri: "direct:objectDetection"
    steps:
      - process:
          ref: "objectDetector"
      - to: "direct:display"

Testing the Example

1. The example will access your default camera (usually /dev/video0)
2. Processed frames will be displayed in a window
3. Detected objects will be highlighted with bounding boxes

Performance Considerations

• Processing resolution impacts performance
• Consider frame skipping for better performance
• Use hardware acceleration when available (CUDA, OpenCL)

Next Steps

• Add support for multiple cameras
• Implement motion detection
• Add video recording capabilities
• Integrate with cloud storage for processed videos