How does AI face analysis work?
Aura Mirror's AI examines your photo for facial action units — the standardized muscle movements of Ekman's Facial Action Coding System — plus structural features and apparent state. A deterministic scoring layer then converts that evidence into seven reproducible presence scores, and the reading explains each one with the evidence it saw.
What are facial action units?
Action units (AUs) are the individual muscle movements a face can make — AU 4 is the brow lowerer, AU 12 the lip-corner puller of a smile. FACS, the coding system behind them, is the standard taxonomy used in behavioral research. Your dossier lists the AUs detected on your face and what they signal together.
Can AI really read emotions from a face?
It can read expression — the visible configuration of your face — and report what that configuration typically signals to observers. Whether you feel that emotion inside is yours to know. Aura Mirror deliberately frames its output as 'what you project,' because that's what a photo actually contains.
How is this different from a beauty-score app?
Beauty scorers reduce your face to one number and no information. Aura Mirror reads expression, state, and structure across seven dimensions with cited evidence, and ends in actions — what to adjust, not a verdict to internalize.
Is my photo used to train AI?
No. Your photo is processed to generate your reading and stored only in your private account history, which you can delete anytime. It is never sold, never used for advertising.