OUR METHOD

Useful context.
Clear limits.

We use automated analysis to map features, then help you think through your options.

What happens to your photo

You choose a photo and consent to processing. Your browser loads a copy of Google’s MediaPipe Face Landmarker model hosted with this site. The model detects facial landmarks locally. Your image, landmark coordinates, and report are not transmitted by our application.

What the analysis does

The face detector and landmark mapper are machine-learning models. Simple calculations describe visible proportions and camera alignment. Recommendation text comes from a curated library selected by your goal, not a medical model or a clinician. We do not describe generic care suggestions as a diagnosis based on your skin.

No ideal-face score

Measurements describe the image, not your worth, health, attractiveness, or an ideal face. Camera distance, focal length, expression, and head position change apparent proportions. Small differences are not medical findings.

What we don’t assess

We do not identify people, infer ethnicity, health status, or other sensitive traits, diagnose skin conditions, predict outcomes, prescribe treatment, or determine suitability for surgery, injectables, or other procedures.

When a photo isn’t enough

If no face, multiple faces, or a strongly turned face is detected, we ask for a different image. Even a good photo cannot substitute for examination, medical history, or a qualified professional’s judgment.

Technology reference

Google MediaPipe Face Landmarker documentation ↗. Google does not endorse this site. The model and runtime are served from our site; we do not send your photo to Google.