justadudewhohacks/face-api.js — reverse-engineered prompt
Reverse engineered prompt
Build me a JavaScript face detection and recognition project using face-api.js.
I want it to run in the browser and let a user open a page, use their webcam or upload an image, then see faces detected with boxes drawn over them. It should also show face landmarks and basic predictions like expression, age, and gender when the right models are loaded. Please include the model loading flow, use the weights from the project or a public models folder, and make the UI simple enough that someone can tell what is happening.
Also include a small Node.js example that can run face detection on an image file from disk, since this library supports both browser and Node. Keep the code clean and easy to follow, with clear setup instructions for installing dependencies and running the browser demo and the Node demo. Look up the current docs online if you need to.
Want more depth? Deep Reverse