A LITTLE TRANSPARENCY
What goes into your rating.
The short version
FaceRater detects facial landmarks in your browser with Google MediaPipe. It calculates two-dimensional ratios and alignment measurements from the uploaded photo. MediaPipe does not provide a beauty or attractiveness score.
The experimental rating
Our beta rubric combines paired-landmark reflection error (50%), eye-width difference (25%), and central landmark offset (25%). Each error is normalized to face or eye width. The displayed score is 10 × exp(−4 × weighted error), rounded to one decimal. These weights and the scale are experimental product choices. They have not been validated as measures of attractiveness or against a representative population.
What the measurements mean
Symmetry describes similarity to mirrored landmark positions. Eye widths and corner angles describe what is visible in the photo. Length and width ratios refer to mesh points, not anatomical or hairline measurements. We do not infer age, ethnicity, personality, health or a population percentile.
Why photos give different results
Camera distance, angle, expression, lighting and occlusion can change the geometry. We check for one detectable face, sufficient image and face size, extreme head roll and off-center pose, very poor exposure, and severe softness. These checks do not guarantee a perfect image.
The full report
The free score uses a fixed formula. Paid reports build on your saved measurements with AI-assisted explanations and a practical plan. The 4-Week Report adds 44 descriptive measurements across seven regions and generated styling concepts. None is presented as an ideal proportion. Our public sample uses a synthetic adult photo and is labeled as an example.
Your original score stays put
A report and a share card use the saved analysis snapshot. The report generator cannot replace its score. Method version: photo-balance-beta-1.