AI Side Profile Rating: Convex, Straight, or Concave?

EXAMPLE ANALYSIS: 478 LANDMARKS MAPPED ON A SAMPLE PHOTO, NOT YOURS.
What a reading looks like
Real components · sample dataA starting line, not a verdict.
What we won't show you: a confidence band (not until our test-retest study ships) · a percentile (not until a reference population is published) · a skin score from one photo (lighting lies). Every "not scored" comes with its reason.
Side profiles fall into three classes: convex (retruded chin), straight (orthognathic) and concave (protruded chin), measured as the angle between glabella, subnasale and pogonion.
This page is part of our AI facial analysis, and it explains all three classes in full. Your free scan analyzes a straight-on photo on your device. The side profile angle itself is calculated inside your Attractiveness Report. A class is a starting line, not a verdict.
The Short Version

The free scan is frontal, and it says so. Profile measurements need this view: a level, true side photo where the line from forehead to chin sits in clean silhouette. That photo is taken inside the paid report flow, where the geometry it supports is measured and everything it cannot support is named instead of guessed.
What Is a Side Profile, and What Does It Actually Show?
A side profile is the lateral view of a face, from forehead to chin. It shows projection, the front-to-back position of the forehead, nose, lips and chin, which a straight-on photo cannot capture at any resolution.
The landmarks read down a profile in order: glabella, nasion, pronasale, subnasale, the upper and lower lip, pogonion, menton and gonion, from brow to jaw corner. A side profile class is not a face shape. Face shape describes your outline from the front, oval, round, square and the rest. Profile class describes projection from the side, and a face carries any combination of the two.
What Is a Convex Profile?
A convex profile is one where the chin sits behind the line running from the forehead to the base of the nose, producing a larger convexity angle. Convex profiles are sometimes called a retruded chin, a recessed chin or a weak chin, three everyday names for the same class, not a diagnosis.
What Is a Straight Profile?
A straight profile is one where the forehead, nose base and chin sit close to a single line, sometimes called orthognathic, meaning a jaw and forehead in normal alignment. This class sits near the middle of the convexity range described on this page, between the convex and concave classes on either side.
What Is a Concave Profile?
A concave profile is one where the chin sits ahead of the forehead-to-nose-base line, producing a smaller convexity angle. It is sometimes called a protruded chin. The everyday word flat is also used for this class, and what that word actually means is answered later in the FAQ below.
How Your Profile Angle Is Measured, in Degrees
Your profile angle is the angle formed at the subnasale between two lines: one running to the glabella, one running to the pogonion. A smaller angle reads as more convex, one close to a straight line reads as straight, and a larger angle reads as concave. The same photo also produces four sub-angles, the nasolabial, nasofrontal, mentolabial and submental-cervical angles. It also yields chin projection, measured against the Ricketts E-line as a ratio rather than millimeters, because an uncalibrated photo has no scale. Your class, your angle and the sub-angles are calculated automatically inside your Attractiveness Report, each one listed with the weight it carries in your score.
- Take a true side-on photo with a rear camera held at eye level.
- Stand at least two meters back and zoom the frame in, rather than moving closer.
- Keep your chin in its natural resting position and your head level.
- Mark the glabella, the base of the nose and the chin, then measure the angle between the two lines with a protractor or a photo-editing tool.
A photograph reads the soft tissue on the surface of a face. A cephalogram, an X-ray taken in a clinic, reads the bone underneath, and the two are not interchangeable. The same three landmarks also feed a golden ratio comparison, a separate and more contested measurement.
What Is a Normal Profile Angle? Reference Values by Population and Sex
There is no single normal profile angle. Published photogrammetric studies report different average convexity angles for different populations, and a class read against one population's values is not the same read against another's.
| Studied population | What the study measured | Source |
|---|---|---|
| Himachal population, India [10] | Soft-tissue facial profile by photogrammetric analysis | Bhandari et al., 2015 |
| African American adults, Philadelphia [11] | Cephalometric norms in a normal-occlusion sample | Kim et al., 2025 |
| Japanese-Brazilian adults [12] | Facial profile in adults with normal occlusion and well-balanced faces | Scavone et al., 2006 |
| Kenyan and Chinese adults [4] | Angular photogrammetric comparison across two populations | Wamalwa et al., 2011 |
Sex is reported the same way in the literature. Several of these studies measure male and female samples separately rather than pooling them, because the average convexity angle differs by sex as well as by population. This is also why your Attractiveness Report names the reference population behind any figure it quotes, rather than ranking you against our own visitors, who are not a representative sample of anyone's face. No percentile is published until a reference population exists that supports one.
Every class and score on this site is explained the same way: see what your score means.
Why Your Camera Changes Your Side Profile
A photo taken close to the face exaggerates how far the nose and chin project, a distortion measured directly in photogrammetric research on short-distance portraits [5]. Head position changes the reading too. Tilting the chin up reduces the measured convexity angle, and tilting it down increases it, an effect measured on cephalometric images of the soft-tissue profile [2][6]. The manual protocol above calls for eye-level framing and a two-meter working distance for exactly this reason. An automated measurement in your Attractiveness Report declines to score a photo taken too close or with the head tilted out of a natural position.
What the Evidence Says About Side Profile and Attractiveness
Soft-tissue profile measurements predict how attractive a face is rated, using machine learning models trained on cephalometric data [7]. The prediction is statistical, built from many faces at once, not a rule about any single face.
How a straight or convex profile is judged for the lips and chin depends on who is rating it. Orthodontists, oral surgeons and laypeople rating the same Persian facial profiles disagreed on which read as most attractive [8]. A profile that reads well to one group can read differently to another.
The profile is one input among several that predict how a face is judged for beauty, alongside factors unrelated to convexity [9]. Convexity contributes to the judgment; it does not decide it alone.
Laypeople and clinicians also disagree on convex profiles specifically: what a clinician classifies as treatment-relevant is not always what an untrained rater calls unattractive [15]. That gap bounds all three findings above. A measured class is a clinical or statistical category, not a lay verdict.
Is a Convex or Concave Profile Bad?
Neither convex nor concave is bad. Both are ordinary anatomical variation, and which class a given angle falls into depends on which population's reference values it is measured against [4]. A comparison of Kenyan and Chinese adults found convexity angles differ measurably between the two groups [4]. A profile read as convex against one population's average can read as closer to typical against another's. This page reports a class against a named reference population rather than one universal number, because a universal number would misrepresent that variation as a defect. The research behind these classes comes from rating tasks on photographs and altered images, not a rule about which faces belong to which people, and that distinction matters more than the label itself.
Why your side profile looks worse to you than to everyone else
You see your face reversed in every mirror you use, so a photograph, which is not reversed, shows an arrangement you rarely see. A study of people who had already sought plastic surgery consultations found they preferred the mirror-reversed version of their own photographed face over the true, unreversed image [13]. That preference was measured in a specific clinical sample seeking cosmetic consultation, not the general population, so it explains a common feeling rather than proving your side profile looks worse than it does.
What This Score Cannot Tell You
This page returns a class, and in your Attractiveness Report an angle in degrees, not a single score. Here is what neither one can tell you, and it is where the account of the measurement ends.
The full method and the camera-distance estimate are on how we measure, and where the measurement fails. What you do with the class you just read is a separate question, and it is the one most people came here to answer.
You Know Your Class: What Should You Do With It?
A convex, straight or concave class is a starting line, not a verdict. Before it means anything, confirm it is not a camera artifact. Then separate what a profile genuinely changes about how your face reads from what it does not change at all.
How to confirm it
- Retake the photo with the rear camera at eye level.
- Stand two meters back and zoom the frame in rather than moving closer.
- Keep your chin in its natural resting position and match the lighting from the first photo.
- Compare the two angles, not the two class labels.
A small difference between two careful retakes is ordinary camera noise, not a change in your bone structure. A difference that moves you across a class boundary on two otherwise identical photos usually means one of the two broke the protocol above.
What it changes
The same photo that produced your profile angle also contains your jaw-to-neck line, the submental-cervical angle, one of the direct inputs to how defined your jawline actually is. Forward head posture, camera distance, hair and beard framing, and photo lighting all change how a profile reads in a picture. None of them move the bone underneath.
What it does not change
Posture, lighting and camera distance change how a profile photograph looks. They do not change the bone and cartilage underneath, which no photo can alter. Mewing is a claim about reshaping that bone through tongue posture, graded elsewhere on evidence, not demonstrated here. Orthognathic surgery and genioplasty are real operations that change the bone directly, and they belong with a qualified clinician after an in-person examination, not a photo tool. The term looksmaxxing side profile names an online self-improvement subculture built around this exact measurement, defined here and not adopted.
Side Profile and the Rest of the Face: What Else Is Measured
Side profile is one component in a larger analysis. The full AI face analysis scores every component, symmetry, proportions, harmony and this profile classification, from the same set of landmarks in one pass.
For a full 1 to 10 rating from a straight-on photo instead of a lateral one, the full attractiveness test is the namesake instrument on this site. To rate a front-facing photo instead of a profile one, that tool covers the front view. This page covers the side view, and the two report different things about the same face.
Is the Side Profile Rating Accurate?
The measurement is reproducible: the same photo, read by the same version, returns the same angle every time, a narrower claim than being correct about your face. Photogrammetric measurement carries a documented uncertainty range against direct and radiographic measurement of the same landmarks [14]. Part of managing that uncertainty is refusing photos that would make it worse, a front-facing shot, a three-quarter turn, one taken from arm's length, rather than scoring them anyway. Refusing a bad photo is part of the accuracy claim, not an exception to it.
More Face Measurements
- The face symmetry test measures how closely the left and right sides of your face match.
- The golden ratio calculator checks your proportions against the contested golden ratio, with the evidence stated plainly.
- The facial harmony test scores how your features work together as a whole.
Not medical advice. This is a measurement of a photo, not a diagnosis, a clinical assessment, or a verdict on you.
It measures geometry, not worth. The number tells you where you start. It never tells you what you are.
on every page
of this site
Your report, in full.
- Side profile
- Skin, measured properly
- Averageness
- Dimorphism
- Ranked plan
Paid photo deleted within 24 hours · never used for training
- 40+ measurements, 7 regionsPP. 2-9
- Side profile, from a second photoPP. 10-13
- What moved your score, rankedPP. 14-16
- Four week plan and day 30 re-scanPP. 17-20
Frequently Asked Questions
What makes a side profile attractive?
No single feature makes a side profile attractive. Predictive models trained on cephalometric measurements link it to a combination of soft-tissue values read together, not one angle in isolation [7][9]. Findings from a small set of studies do not translate into a rule about any one face.
How do I check if my side profile is good?
Take a true side-on photo two meters back with the frame zoomed in, mark your forehead, nose base and chin, then measure the angle between them, the method described above. Your Attractiveness Report runs the same measurement automatically and states the weight it carries in your score.
What causes a bad side profile?
Convexity is mostly a matter of underlying skeletal and soft-tissue proportion, not a fixable cause, and it varies measurably by the population studied [4]. A convex or concave reading describes anatomy, not a defect to diagnose.
Do most people hate their side profile?
Many people find their own side profile photo unfamiliar because mirrors show a reversed image they have seen far more often than the true, unreversed one [13]. That finding comes from a specific clinical sample, not a universal law about faces.
Can AI rate my side profile accurately?
An AI measurement of the same photo and version returns the same angle every time, which is what reproducible means, and it is a narrower claim than being correct about your face. Accuracy also depends on declining photos taken from the wrong angle or distance.
How do I fix a bad side profile?
Nothing done at home changes the bone or cartilage that sets your convexity class. Posture, camera distance and photo framing change how a profile reads in a picture without touching the skeleton underneath.
Does my face shape affect my side profile class?
No, face shape and side profile class measure different views of the same face. Face shape describes your outline from the front; profile class describes projection from the side, and a face carries any combination of the two.
Is a flat side profile the same as a bad one?
No, flat is an everyday word for a lower convexity angle, and whether that angle reads as convex, straight or concave depends on which population's reference values it is measured against [4]. Flat describes a class, not a defect.
Does my photo get uploaded when I measure my side profile?
No, your free scan analyzes your photo on your device, and nothing is uploaded. See the privacy page for exactly what is and is not transmitted. If you buy your Attractiveness Report, that photo is deleted within 24 hours and never used to train a model.
References
Every claim above, traceable
- Mirmohammadsadeghi H, Vazirizadeh Y, Rokni H, et al., Journal of Imaging Informatics in Medicine 2026 - deep learning classification of facial profile convexity from profile photographs in orthodontics, the basis for the convex, straight and concave classification and angle used on this page.
- Wozniak K, Piatkowska D, Lipski M, Advances in Clinical and Experimental Medicine 2012 - the influence of natural head position on the assessment of facial morphology.
- Anic-Milosevic S, Lapter-Varga M, Slaj M, European Journal of Orthodontics 2008 - analysis of the soft tissue facial profile by means of angular measurements, defining the nasolabial, nasofrontal, mentolabial and E-line sub-angles.
- Wamalwa P, Amisi SK, Wang Y, et al., Journal of Craniofacial Surgery 2011 - angular photogrammetric comparison of the soft-tissue facial profile of Kenyans and Chinese adults.
- Ward B, Ward M, Fried O, Paskhover B, JAMA Facial Plastic Surgery 2018 - nasal distortion in short-distance photographs, the selfie effect.
- Hoogeveen RC, Sanderink GC, Berkhout WE, Dentomaxillofacial Radiology 2013 - effect of head position on cephalometric evaluation of the soft-tissue facial profile.
- Pirayesh Z, Sohrabniya F, Ourang SA, et al., International Dental Journal 2026 - predicting perceived profile attractiveness from cephalometric measurements using machine learning.
- Imani MM, Sanei E, Niaki EA, et al., American Journal of Orthodontics and Dentofacial Orthopedics 2018 - esthetic preferences of orthodontists, oral surgeons and laypersons for Persian facial profiles.
- Ghorbanyjavadpour F, Rakhshan V, American Journal of Orthodontics and Dentofacial Orthopedics 2019 - factors associated with the beauty of the soft-tissue profile.
- Bhandari V, Singla A, Mahajan V, et al., Indian Journal of Dental Research 2015 - soft tissue facial profile in the Himachal population, a photogrammetric analysis.
- Kim LEH, Jeong J, Setzer FC, et al., Journal of Orofacial Orthopedics 2025 - cephalometric norms for African Americans with normal occlusion in the Greater Philadelphia region.
- Scavone H Jr, Trevisan H Jr, Garib DG, et al., American Journal of Orthodontics and Dentofacial Orthopedics 2006 - facial profile evaluation in Japanese-Brazilian adults with normal occlusions and well-balanced faces.
- de Runz A, Boccara D, Chaouat M, et al., Journal of Plastic, Reconstructive and Aesthetic Surgery 2016 - female plastic surgery patients prefer mirror-reversed photographs of themselves, a prospective study.
- Fatima F, Hallolli C, Tubaki R, et al., Cureus 2025 - comparative evaluation of photogrammetric, radiographic and direct measurements in facial analysis.
- Maple JR, Vig KW, Beck FM, et al., American Journal of Orthodontics and Dentofacial Orthopedics 2005 - a comparison of providers' and consumers' perceptions of facial-profile attractiveness.