JEOresearch/EyeTracker — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a simple Python eye tracking tool that can find and follow a pupil in an eye video. I want to be able to run it from the command line, pick a video file if one is not already set, and see the result with an ellipse drawn over the pupil as the video plays.

Please include a small test video flow, clear setup steps, and keep it lightweight using OpenCV and NumPy. It should work best with normal 640 by 480 eye footage, crop the image sensibly if needed, and avoid dark borders that could confuse the tracking. Add a faster lite mode too, even if it is a little less accurate and needs good lighting.

Make the main processing easy to reuse later with a webcam or single image, and return the pupil ellipse data from the tracking function. Include notes about the NumPy version issue and basic troubleshooting for lighting and camera placement. Look up current docs online if you need to.

Want more depth? Deep Reverse