Free tool · runs in your browser
Face Symmetry Test
Upload a photo. Symmi maps 478 landmarks, corrects for head tilt, and scores how far each feature sits from your midline. You get a symmetry index and eight readings, not a vibe.
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.
How it is measured
Distance from the midline, feature by feature.
The midline is drawn from the glabella (between the brows) through the chin. Every paired feature is measured on both sides and the difference is expressed as a fraction of face width, so the score does not depend on photo size. The index weights eyes and mouth more than cheeks because that is where people notice asymmetry first.
- Symmetry index
- 100 minus the weighted mean deviation. 92 and above reads as symmetric to most viewers; 86 to 91 is mild; under 86 is visible at a glance.
- Chin and nose offset
- How far the chin point and nose tip sit left or right of the brow midline (the vertical through the point between the brows, after head tilt is removed), as a percent of face width. Under 1.5 percent is usually unnoticed.
- Eye, brow, mouth height
- Vertical difference between the left and right eye centre, brow peak and mouth corner.
- Eye width
- Difference in horizontal eye opening, left versus right.
- Head roll
- The tilt the tool removed before measuring. Over 3° means retake the photo for a cleaner read.
Measure it yourself
How to check your face symmetry yourself
Light and camera angle create more fake asymmetry than faces do, so start with the photo.
- 1Take a straight-on photo at eye level from at least an arm's length away, with even light from the front and a neutral face. Side light and close-up selfies exaggerate differences.
- 2Draw the midline: a vertical line through the point between your brows, at right angles to the line through your pupils.
- 3Check heights: do both pupils, both brow peaks and both mouth corners sit on the same horizontal line?
- 4Check distances: compare how far each eye corner, nostril, mouth corner and jaw corner sits from the midline on each side.
- 5Express each difference as a percent of your face width. Differences under about 1.5 percent usually go unnoticed.
Skip the ruler: the free tool above does it from one photo in about 30 seconds ↑
What's normal
How much asymmetry is noticeable?
Studies of what viewers notice put the thresholds at about 2 mm for the eyelids, 3 mm for the mouth corners, 3.5 mm for the brows, 4 mm for the nose tip and 6 mm for the chin (Wang et al., 2017). Almost nobody measures perfectly symmetric.
- Symmetry index (this tool)
- Usually unnoticed: 92 and above
- Noticed: 86 to 91 mild; under 86 visible at a glance
- Eye height difference
- Usually unnoticed: Under 1% of face width (about 2 mm)
- Noticed: Above 1%
- Brow height difference
- Usually unnoticed: Under 1.5% of face width
- Noticed: Above 1.5%
- Mouth corner height
- Usually unnoticed: Under 1.5% of face width
- Noticed: Above 1.5%
- Chin or nose offset
- Usually unnoticed: Under 1.5% of face width
- Noticed: Above 1.5%
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.
Wang TT, Wessels L, Hussain G, Merten S (2017). Discriminative thresholds in facial asymmetry: a review of the literature. Aesthetic Surgery Journal. PubMed 28200081
Review of how many millimetres of asymmetry people can actually notice, feature by feature.
- 2.
Grammer K, Thornhill R (1994). Human (Homo sapiens) facial attractiveness and sexual selection: the role of symmetry and averageness. Journal of Comparative Psychology. PubMed 7924253
Early study linking facial symmetry and averageness to attractiveness ratings.
- 3.
Rhodes G (2006). The evolutionary psychology of facial beauty. Annual Review of Psychology. PubMed 16318594
Review of what the evidence does and does not show about symmetry and facial beauty.
- 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.
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
Is anyone perfectly symmetric?+
No. Almost every measured face scores under 100. The question is whether the difference is above the roughly 2 to 3 percent that viewers can see. This tool tells you which feature carries the difference.
Why does my score change between photos?+
Head tilt, a turned head, wide-angle lenses and uneven light all shift the landmarks. Use a straight-on photo at arm's length or further, and compare photos taken the same way.
Which asymmetries can training change?+
Soft tissue: a lower mouth corner, a droopier brow, a puffier cheek. Bone, like a nose that sits off centre, does not move. The Symmi app splits every reading into bone and trainable and builds the routine from the trainable ones.
Is my photo uploaded?+
No. The model downloads into your browser and runs there. The photo never leaves your device and we never see it.
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.