Ford Vision Alignment

A vision-based alignment system for operator UI with real-time camera feed and alignment controls.

Setup

1. Python Environment

Create the conda environment from environment.yml:

conda env create -f environment.yml
conda activate vision

2. Frontend Dependencies

Install Node.js dependencies:

npm install

3. Camera Configuration

Add your RTSP camera stream URL to .env.local:

VITE_LINE1_MAIN_STREAM_URL=rtsp://YOUR_CAMERA_IP:8554/live.sdp

Running the Application

The application requires three servers running simultaneously:

1. Camera Server

conda activate vision
python backend/start_server.py

Runs on http://localhost:8000

2. Region Calibration Server

conda activate vision
python backend/region_calibration_server.py

Runs on http://localhost:5001

3. Frontend Development Server

npm run dev

Runs on http://localhost:3000

Access the application at: http://localhost:3000/tasks

Usage

• Navigate to the main dashboard to view camera feeds and alignment controls
• Use the alignment view to see real-time alignment data and instructions
• Access demo view at /demo route for testing