Tools
Free face analysis tools
Explore one feature at a time with a free browser tool: face shape, eye proportions, jawline, skin tone and more. For a complete assessment and personal improvement plan, start with the full face scan. Tool photos stay on your device.
See the full picture with a face scan →
27 tools
Which face analysis tool should I use?
Pick the card for your question above, or start from what you are deciding:
- Haircut, frames or colours: face shape, beard style, eyebrow shape, skin tone finder, visual weight test and pupillary distance for ordering glasses.
- Eyes: canthal tilt, hunter eyes, eye shape, eye spacing and the eye type test with its fox, siren, doe, prey, almond, upturned and downturned pages.
- Proportions: symmetry, golden ratio, facial ratios, facial thirds, fWHR, facial harmony, the jawline test and the lip shape test. To see your face the way others do, the true mirror.
- The whole face at once, with a plan: the full rating.
Every geometry tool on this page works the same way. A face mesh runs on your device, finds a few hundred landmarks on the photo, and reads one measurement family from them: widths for face shape, mirrored pairs for symmetry, corner heights for canthal tilt, and so on. The two colour tools read pixels instead: the skin tone finder samples the colour of your cheeks and forehead, and the hair colour try-on uses a small hair model to recolour only your hair. The glasses try-on uses the same face mesh to fit frames instead of reporting a number. The arithmetic is simple and it is all done locally. Nothing is sent to a server, nothing is stored, and the photo is gone when you close the tab. That is also the limit: a single-metric tool sees one thing and cannot see the rest of the face, so each page says what the number means, and what it does not.
Each tool reports a band first and the raw reading beside it, and the band is the part to trust, because a photo cannot support the last decimal. Head pose, lens distance and light move every one of these readings, and the photo rules on each page exist to keep them honest. Take the photo at eye level, at arm's length or further, face square to the lens, in even front light, and the readings hold across repeat photos; break the rules and they drift by a band.
What does the full rating add?
The tools are single measurements. The AI face rating is the whole picture: one photo, a score from 0 to 100 on a scale where an average face lands in the middle, a tier, the PSL equivalent, and, in the report, visible traits across harmony, eye area, jaw, skin and hair, each tagged workable or structural with a capped, non-surgical potential. It also reads the things no landmark tool can: skin evenness, under-eye condition, hair density, how the jaw reads at your body composition. The rules it follows are on the methodology page, and every term the tools use is defined in the glossary.
Questions
Do the face analysis tools send my photo anywhere?
No. Each tool runs a face mesh in your browser, does its arithmetic on your device and shows the result. No photo, landmark or result is sent to BecomeTen or anyone else, and nothing is kept after you close the page. The full rating is different, because a model has to see the photo to rate it, and that page explains how it handles the image.
Can I check how the tools compute their numbers?
Yes. The measurement code behind every geometry tool on this page is published as open-source code in the GitHub repository MartinSaraka/face-metrics under the MIT licence (the GitHub link is in the footer of every page). It contains the landmark indices, the roll correction, the pose gates and the unit tests, so anyone can read exactly how a face shape, a symmetry score or a canthal tilt is derived, or reuse the functions in their own project.
Are the browser tools as accurate as the paid report?
They answer different questions. A tool reads one proportion from landmarks and is as accurate as the landmarks and the photo allow, which is why each reports a band. The report is a rated opinion across dozens of traits against a fixed rubric, including things landmarks cannot see. Neither is a measurement of your face; both say so.
Why do the tools lead with bands rather than exact figures?
Because a photo is a flat projection and the apparent size of everything on it moves with head pose, lens distance and light, usually by more than real faces differ. An exact figure would mostly describe the camera. Bands are the resolution a photo can carry honestly. Where a tool shows a figure, it sits beside the band for comparing your own photos; pupillary distance is the one meant to be used as a figure, and even that needs an optician to confirm it.