Face Rating: Measure Your Facial Ratios
Upload one front-facing photo and get an overall score built from four measured parts — symmetry, facial thirds, facial fifths and golden-ratio proportions — with the formula and the landmarks behind every number.
- Photo never leaves your device
- Every formula published
- Same photo, same result
- 1Add a photo
- 2See your ratios free
- 3Full report $9.90 · optional
Use a straight-on, evenly lit photo
Hair off the forehead, neutral expression, no glasses.
Analysed on your device. Your photo never leaves your device.
What the tool sees, and what you get
Below is MediaPipe’s reference face — the average mesh the landmark model is built on, not a real person — measured by this site’s engine. Your photo gets the same lines, numbers and card.
Overall score
Band A · 87–100
Measured proportions sit very close to every reference value.
-
Symmetry 35%100
-
Facial thirds 25%92.5
-
Facial fifths 25%81.4
-
Golden ratio 15%84.9
Facial thirds
Reference 18 / 41 / 41
Face shape
Oval
- Oval83.4%
- Round6.6%
- Oblong3.3%
Photo confidence 92.5%. The reference mesh is an exact mirror, so its symmetry is 100 by construction; real photos differ.
Measure my face — freeSix free tools, one photo
Pricing
Every tool on this site is free to use. If you want everything in one place, there is one paid option.
Free tools — $0
Overall score, symmetry, face shape, golden ratio and jawline analysis, each with its formula. No sign-up, no limit.
Full Facial Analysis Report — $9.90 one-time
- All 58 measurements with the landmarks used
- Every score with its formula, golden ratio deviation table, symmetry by region
- Face shape match across seven shapes, annotated face map, written notes
- PDF report generated on your device
Single payment of $9.90 USD (plus any sales tax shown at checkout). No subscription, no automatic renewal, no account. Paid through Waffo, our merchant of record. See the refund policy and terms.
At a glance
- What it is
- A free browser tool that measures facial proportions from one front-facing photo: symmetry, facial thirds and fifths, golden ratio, face shape and jawline, in seconds.
- Input
- One front-facing photo, uploaded or taken with the camera. Side profiles are not measured; items that need one are marked “Not scored — needs a side photo”.
- Free
- All six tools, unlimited, no sign-up.
- Price
- Full Facial Analysis Report: $9.90 one-time. No subscription, no automatic renewal.
- Account
- None needed.
- Privacy
- The photo is analysed in your browser and never uploaded.
- Method
- MediaPipe Face Landmarker (478 landmarks). Every formula and constant is published; the same photo always gives the same result.
- Report includes
- 58 measurements, all scores with formulas, golden ratio deviation table, symmetry by region, match across 7 face shapes, annotated face map, PDF made on your device.
- Does not do
- No AI image generation, no face editing, no comparing you with other people or famous faces.
- Refunds
- See the refund policy.
What your result shows
This is a free face rating that measures rather than guesses. When you add a photo, a face-landmark model places 478 points on your face inside your browser. The tool then reads distances and angles between those points and reports four things in one card.
Overall score and band
A number from 0 to 100 and a band from A to E. The band only describes how close your ratios are to the reference values; it does not compare you with anyone.
Four part scores
Symmetry, facial thirds, facial fifths and golden ratio, each with its own bar, the raw measurement behind it and the reference it is compared with.
Confidence and photo quality
A confidence value with one decimal (for example 72.6%) and plain notes on head angle, light and the hairline, so you know how far to trust this particular photo.
You also get an annotated copy of your photo: yellow lines mark the facial thirds, teal lines the facial fifths and red dots the jaw corners and chin. It is drawn on your device and is never uploaded. If the head is turned more than 10° or tilted too far up or down, the card asks for a retake and withholds the scores, because a tilted head changes the ratios more than any real difference between faces would.
What is measured, and where
Every length is divided by the distance between your pupils (iris centres, landmarks 468 and 473), so the size of the photo does not matter. Before measuring, the eye line is levelled and a small head turn is undone using the model’s depth estimate. These are the measurements that feed the score:
| Measurement | Landmarks | How it is calculated |
|---|---|---|
| Symmetry (mirror error) | 28 midline points, 220 left/right pairs | A midline is fitted through the centre of the face; each left point is mirrored across it and its distance to the matching right point is averaged, as % of pupil distance. |
| Upper third | 10 → 9 | Upper forehead to the point between the brows. Estimated: the model cannot see the hairline. |
| Middle third | 9 → 2 | Between the brows to the base of the nose. |
| Lower third | 2 → 152 | Base of the nose to the bottom of the chin. |
| Facial fifths | 234, 33, 133, 362, 263, 454 | Face width split at the outer and inner eye corners into five segments, each as % of face width. |
| Mouth ÷ nose width | 61↔291 ÷ 64↔294 | Mouth corners against the outer edges of the nostrils, compared with φ = 1.618. |
| Nose-base-to-chin ÷ lips-to-chin | 2, 13/14, 152 | Two vertical spans of the lower face, compared with φ. |
| Lower ÷ upper lip | 0, 13, 14, 17 | Height of the lower lip against the upper lip, compared with φ. |
The full report adds the rest of the 58 measurements — eye widths and spacing, canthal tilt, brow height, nose and mouth proportions, jaw and chin widths, projected jaw angle and seven-way face shape. A few things cannot be measured from a front photo at all; the jaw’s true gonial angle and chin projection are listed as Not scored — needs a side photo instead of being filled with a guess.
How the overall score is built
Each part starts at 100 and loses points in proportion to its deviation from the reference. The four parts are then weighted and added. Nothing is hidden: the constants below are the ones the tool runs.
| Part | Weight | Reference | Points lost |
|---|---|---|---|
| Symmetry | 35% | 0% mirror error | 6 per 1% of pupil distance |
| Facial thirds | 25% | 18 / 41 / 41% | 2.5 per percentage point, summed over the three thirds |
| Facial fifths | 25% | 20% each | 1.5 per percentage point, summed over the five fifths |
| Golden ratio | 15% | φ = 1.618 | 1.5 per 1% mean deviation over three ratios |
Why the thirds reference is 18/41/41, not 33/33/33. The classic “equal thirds” start at the hairline. The model cannot see the hairline, so the upper third starts at the upper forehead, which makes it shorter. On MediaPipe’s own average face model the same landmarks give 19.1 / 39.5 / 41.4. Across the 13 front-facing portraits we test with — a small sample, not a population — the medians were 20.4 / 38.7 / 40.7. Ordinary faces therefore do not reach 100 on thirds, and that is expected.
How to read your score
The bands describe the measurements in this photo. They are not a ranking: we have no database of other people’s faces and never compare yours with anyone else’s.
| Band | Score | What it means |
|---|---|---|
| A | 87–100 | Measured proportions sit very close to every reference value. |
| B | 83–86 | Close to the reference values, with one or two small deviations. |
| C | 78–82 | Close on most measures, with a few noticeable deviations. |
| D | 60–77 | Several measures differ from the reference values — typical of everyday faces and phone photos. |
| E | 0–59 | Large deviations from the reference values — often caused by pose, lens distance or expression. A retake usually helps. |
Look at the part scores before the total. A low symmetry part with a high confidence value usually points to a real left–right difference; a low symmetry part with a low confidence value usually points to the photo. The thirds part is always marked estimated because of the hairline. A face rating from a single photo is a snapshot of that photo’s geometry — a second photo taken under the same conditions is the best way to check whether a number is about you or about the camera.
Getting a measurement you can trust
Hold the camera at eye level
A camera below the chin lengthens the lower third and shortens the forehead; one above does the opposite. Eye level keeps the thirds honest.
Step back, then zoom
A phone held close widens the nose and narrows the face edges (lens distortion). Standing about an arm’s length or more away and cropping gives truer widths.
Face soft, even light
Hard side light hides one half of the face and moves the landmarks the model places there, which shows up as false asymmetry.
Keep a neutral expression
Smiling lifts the cheeks, narrows the eyes and widens the mouth, changing the golden-ratio and fifths measurements.
Pull hair off the forehead
Hair over the forehead or jaw edge can shift the outline points the face width and jaw are measured from.
Compare like with like
If you track changes, use the same spot, light and distance each time. Differences under a few points are usually photo-to-photo variation.
Want one measurement in depth? The symmetry analysis breaks mirror error down by region, the golden ratio calculator lists each ratio against φ, the face shape detector shows its decision table, and the jawline analysis explains what a front photo can and cannot show.
Frequently asked questions
How is the face score calculated?
Four parts are measured from 478 landmarks and combined: symmetry 35%, facial thirds 25%, facial fifths 25% and golden-ratio proportions 15%. Each part starts at 100 and loses points in proportion to how far your measurement sits from its reference value. Every constant is published in the method section of this page.
Is the AI face rating accurate?
It is precise about geometry: the same photo always gives the same numbers. It is not a measure of attractiveness. Research reviews (Rhodes 2006; Langlois et al. 2000) show that proportions explain only part of how faces are perceived, so treat the score as a description of ratios in one photo, not a verdict.
What is a good face rating?
There is no “good” or “bad” here. The score tells you how close the measured ratios in this photo are to the reference values, grouped into bands A (87–100) to E (0–59). Band D, 60–77, covers many everyday faces and phone photos. We do not compare you with other people.
Why do different photos of me get different face scores?
Landmarks move with the photo, not just the face. A slight head turn, a phone held close to the face (lens distortion), uneven light or a smile all shift the points that ratios are built from. For comparable results, use the same distance, eye-level camera, even light and a neutral expression.
Is my photo private?
Yes. The face model is downloaded to your browser and runs there; the image is never uploaded. The site has no server-side code that could receive it, and its security policy only allows the page to contact this site and a cookieless page-view counter. You can confirm this in your browser’s network panel.
Does a low golden ratio score mean I’m not attractive?
No. The golden ratio is a mathematical reference, not a beauty standard. Holland (2008) found the well-known φ mask does not describe a wide range of faces, and Pallett et al. (2010) found preferred proportions that differ from φ. The golden-ratio part carries the lowest weight (15%) for that reason.
Method and references
Landmarks come from Google’s MediaPipe Face Landmarker (478 points including the irises), run on your device in single-image mode with the CPU, so the same photo always produces the same numbers. Left/right landmark pairs are derived from MediaPipe’s canonical face model, which is an exact mirror; on that model the tool measures 0% mirror error, 19.1 / 39.5 / 41.4 thirds and 21.0 / 16.9 / 24.2 / 16.9 / 21.0 fifths. The weights (35 / 25 / 25 / 15) follow a common public scoring scheme; the per-part constants are our own and are published above.
Proportions are one ingredient of how faces are perceived, not the whole story. The reviews below summarise what research does and does not support, and the anthropometric studies describe how much normal faces vary.
- Rhodes G. (2006). The evolutionary psychology of facial beauty. Annual Review of Psychology 57:199–226. doi:10.1146/annurev.psych.57.102904.190208
- Langlois JH, Kalakanis L, Rubenstein AJ, et al. (2000). Maxims or myths of beauty? A meta-analytic and theoretical review. Psychological Bulletin 126(3):390–423. doi:10.1037/0033-2909.126.3.390
- Little AC, Jones BC, DeBruine LM. (2011). Facial attractiveness: evolutionary based research. Phil. Trans. R. Soc. B 366:1638–1659. doi:10.1098/rstb.2010.0404
- Farkas LG, Hreczko TA, Kolar JC, Munro IR. (1985). Vertical and horizontal proportions of the face in young adult North American Caucasians: revision of neoclassical canons. Plastic and Reconstructive Surgery 75(3):328–337. doi:10.1097/00006534-198503000-00005
- Farkas LG, Katic MJ, Forrest CR. (2005). International anthropometric study of facial morphology in various ethnic groups/races. Journal of Craniofacial Surgery 16(4):615–646. doi:10.1097/01.scs.0000171847.58031.9e
- Holland E. (2008). Marquardt’s Phi mask: pitfalls of relying on fashion models and the golden ratio to describe a beautiful face. Aesthetic Plastic Surgery 32:200–208. doi:10.1007/s00266-007-9080-z
- Pallett PM, Link S, Lee K. (2010). New “golden” ratios for facial beauty. Vision Research 50(2):149–154. doi:10.1016/j.visres.2009.11.003
Your photo never leaves your device
The face model is downloaded to your browser and runs there. Your image is never sent to our server or any third party — this site has no server-side code or upload endpoint at all, and its security policy lets the page contact only this site and Cloudflare's cookieless page-view counter. Read the privacy policy.