FaceRuler is a measurement instrument with a public scale. We built it because the face-rating world splits into two bad options: black-box AI scoring that can't explain itself, and community forums whose ratings are vibes wearing a lab coat. We wanted a third thing — a ruler: fixed, transparent, and honest about what it can and cannot see.
A face-landmark model marks 478 points on your photo (on your device). Geometric formulas score eight sub-metrics across five dimensions — symmetry, proportions, structure, eyes, and harmony. The weighted overall maps onto the PSL 0–8 scale through a fixed curve anchored at a median of 4.0. The same photo always produces the same score. Reference targets are community values cross-checked against published facial anthropometry (Farkas) and attractiveness research (Rhodes 2006), and recalibrated quarterly from anonymous aggregate data.
Maintained by the FaceRuler team. We treat this like an instrument: when the calibration changes, we announce it in the methodology & sources section of the guide. Found an error in our formulas, references, or wording? Write to us — corrections are credited.
Rating tools can become compulsive, especially for younger users. If checking scores starts affecting your mood, eating, or social life, step away and talk to someone you trust. Support resources: NAMI, Mental Health America.
Questions, corrections, privacy requests: support@faceruler.com — we read everything.