Free tool · runs in your browser

Wide set or close set eyes?

Measure the gap between your inner eye corners and compare it with the average width of your two eyes. See the distances drawn on your photo and understand the one-eye-width comparison.

Drop a photo, paste one, or choose a file

Straight-on, whole face visible, even light. JPG, PNG or WebP.

Runs in your browser. Your photo never leaves your device.

NewSymmi Lab · full face analysis

Your eyes are one reading. Your face has 30.

Your eyes, brows and lashes read in full, the haircut and brow shape that make them stand out, plus your score, your colours and your plan.

  • Your face score and where you rank
  • 4 haircuts tried on your own face
  • Your colours, shades and a step-by-step plan
Get my full analysis · $14.99 →

$14.99 once · no subscription · ready in minutes

Sample before and after: a new haircut and beard
BeforeAfter

Sample result

Understand the comparison

One gap. Two eye widths.

These schematic examples hold both eye widths constant and change only the gap. Real faces can differ in all three distances. The examples are not category boundaries.

Smaller relative gap

A · gapB = average of the two blue widths
0.80×. The gap is 80% of the average eye width.

Equal distances

A · gapB = average of the two blue widths
1.00×. The gap equals the average eye width.

Larger relative gap

A · gapB = average of the two blue widths
1.20×. The gap is 120% of the average eye width.

A reference, not a rule for every face.

The historical orbital canon compares inner-corner distance with eye width. Research documents substantial variation, so equality should not be treated as a universal target. Our ratio describes this photo; it does not diagnose eye spacing or infer anatomy from a population average.

Sources: study of facial canons and systematic review of intercanthal distance.

How it is measured

Inner-corner gap ÷ average eye width.

The gold line joins the inner eye corners. Each blue line joins the inner and outer corners of one eye. All three are straight-line distances in the image; the average of the two blue lengths is the denominator. This ratio is unchanged by geometrically rotating or uniformly resizing the same points, although a new model detection may place those points slightly differently.

Inner-corner gap (A)
Distance between the two inner eye corners. This is different from pupil-to-pupil distance.
Average eye width (B)
Add the photo-left and photo-right corner-to-corner widths, then divide by two. Both individual widths are shown.
Spacing ratio
A ÷ B. 1.00 means the gap equals the average eye width. Lower means a smaller gap relative to eye width; higher means a larger one.
What is not measured
No millimetres, depth, skeletal diagnosis or attractiveness score. This is also different from ESR, which divides pupil span by face width.

Sources

What this is based on.

The measurement and its reference ranges come from published research. Studies use different methods and populations, so compare your number with the range shown on this page rather than with other tools.

  1. 1.

    Bouhadana G, Gornitsky J, Saleh E, et al. (2022). Expanding the classic facial canons: quantifying intercanthal distance in a diverse patient population. Plastic and Reconstructive Surgery Global Open. PubMed 35475286

    Inner-eye-corner distance measured in a diverse population.

  2. 2.

    Al-Sebaei MO (2015). The validity of three neo-classical facial canons in young adults originating from the Arabian Peninsula. Head & Face Medicine. PubMed 25889948

    Tested the 'one eye width apart' canon on real faces.

  3. 3.

    Farkas LG, Katic MJ, Forrest CR, et al. (2005). International anthropometric study of facial morphology in various ethnic groups/races. Journal of Craniofacial Surgery. PubMed 16077306

    Eye-region measurements across ethnic groups.

  4. 4.

    Kartynnik Y, Ablavatski A, Grishchenko I, Grundmann M (2019). Real-time facial surface geometry from monocular video on mobile GPUs. arXiv. arXiv:1907.06724

    The face-mesh model that places the 468 surface landmarks this tool measures from.

  5. 5.

    Ablavatski A, Vakunov A, Grishchenko I, et al. (2020). Real-time pupil tracking from monocular video for digital puppetry. arXiv. arXiv:2006.11341

    The iris model that adds 10 eye landmarks (478 in total), used for pupil-based reference lines.

Our own data: how much each reading moves when the same face is photographed again is published on the methodology page. How we choose sources.

Related tools

FAQ

How do I tell whether my eyes are wide set or close set?+

A common visual comparison looks at the inner-corner gap relative to one eye width. A larger gap is often described as wider-set and a smaller gap as closer-set. This tool reports the ratio using the average width of both eyes, without applying a clinical or beauty cutoff.

Is one eye width the ideal spacing?+

No. The one-eye-width rule is a historical facial proportion convention, not a universal normal range or an attractiveness target. Published measurements show variation across individuals and populations. A ratio above or below 1.00 is not by itself a problem.

Is this the same as interpupillary distance or ESR?+

No. Interpupillary distance runs from pupil centre to pupil centre. ESR in Symmi's facial-ratios tool divides that distance by cheekbone width. This page divides the inner-corner gap by average eye width; these numbers should not be compared as if they used the same formula.

Why use the average of both eyes?+

The two eye widths can differ in a photo. Averaging makes the reference explicit rather than choosing one side, and the individual widths remain visible so you can check them.

Can a photo change the result?+

Yes. Turning your face, perspective, squinting, obscured corners and landmark placement can change the estimate. Use a front-facing photo with an eye-level camera and even light. Roll alone does not change the straight-line geometry, but a different image can change detection.

Can I use this for glasses measurements?+

No. This is a unitless photo proportion and does not provide calibrated pupil distance in millimetres for fitting lenses.

Is my photo uploaded?+

No. Landmark detection, measurement and annotation run in your browser.

One reading here.
Nineteen in the app.

A controlled scan, every reading against its ideal band, bone vs trainable, a plan, and a 3D model on Face ID iPhones.

Symmi Lab · new

Your full face analysis, with 4 haircuts on your own face.

Get my full analysis · $14.99 →