Asian Facial Architecture™

Ask what makes a face work and most answers name a single feature — a jawline, a pair of eyes, a nose. Ask a doctor who assesses faces before treating them, and the honest answer is rarely any one thing. Individual features can be examined in detail, but how a face looks also depends on the relationships between them and on the composition of the whole. Measurements, however precisely taken, do not always explain why a face looks balanced, or does not. Asian Facial Architecture™ is Dr. Gan Lee Ping's framework for understanding these relationships without flattening Asian faces into a single template.

What is Asian Facial Architecture™?

Asian Facial Architecture™ is Dr. Gan Lee Ping's framework for understanding the face as an interconnected three-dimensional structure, in which skeletal framework, soft tissue, proportion, movement, expression and ageing influence one another. It considers the face at multiple scales — from the individual feature and its surrounding region to the face as a whole — because each reveals different information.

The word "architecture" is deliberate. In a building, a structural element can be examined individually while its role is understood in relation to what surrounds it. A face can be considered in much the same way. A cheekbone's apparent projection depends on the fat over the bone and the skin over that fat. A jawline's definition is judged against the chin it meets. A brow's position depends on the muscle holding it and, later in life, on how the eyelid beneath it has changed. Each feature has anatomical and structural characteristics of its own, while its perceived balance is also influenced by the structures around it.

This makes Asian Facial Architecture™ a method of assessment, not a set of target measurements. It draws on established anatomy — skeletal structure, fat compartment anatomy, connective tissue, skin biology — and organises it around two working questions: what is happening within a given feature or region, and how does that sit within the face around it? The structural, layer-by-layer account of how a face ages elsewhere in the Journal describes one part of this picture; Asian Facial Architecture™ is the wider view that holds it alongside proportion, movement, expression and individual variation.

There is no single "ideal Asian face"

There is no single ideal Asian facial proportion, nor does the anatomical literature support reducing Asian faces to one universal template. "Asian" spans a continent, not a phenotype — East Asian, Southeast Asian, South Asian and mixed-heritage faces differ substantially from each other in bone width, nasal projection, eyelid structure and proportion. A framework built around one Asian template starts from the wrong premise before it has looked at an individual face.

This is not only a cultural observation. A systematic review of eleven facial measurements across 2,359 people from twenty-seven ethnic groups — including Korean, Vietnamese, Thai, Japanese and Singaporean Chinese populations — ranked the measurements by their coefficient of variation, calculated within each group and averaged across groups: forehead height was the most variable, and the widths across the cheekbones and outer eyes the least. The same review noted that several earlier studies had found significant differences between the proportions described in the neoclassical canons and the mean proportions of modern non-Caucasian populations. A separate study of 286 Malaysian Chinese, Malay and Indian university students aged 18 to 25 found that 17.1% had a facial index within the golden-ratio range the authors used, with a similar proportion in each of the three groups; that describes conformity to one specific index in that sample, not facial proportion in general.

None of this makes proportion meaningless — the next section makes the opposite case. It means the reference point is the individual face and its own bone structure, not a ratio imported from a population it was never measured on.

The face as architecture

Treat the face as architecture and its parts stop being a list. The skeleton provides the underlying framework: facial width, cheekbone projection and jaw structure establish relationships over which soft tissue is layered. Temple contour, by contrast, depends substantially on the soft tissues overlying bone — an early example of why facial architecture cannot be reduced to the skeleton alone. Soft tissue sits over the frame in layers: deep fat close to bone, muscle and connective tissue above that, superficial fat and skin nearer the surface. The upper, middle and lower thirds of the face are read in relation to each other, not as independently graded zones.

Two consequences follow. A feature can be assessed on its own terms, but its neighbours give that assessment context — a nose that looks large on a narrow face may look proportionate on a wider one. And projection and width interact: a structurally wide face can still read as balanced if projection through the midface and jaw keeps pace with that width, independent of whether either measurement is "large" by any external standard.

Chin and jaw are the clearest example of two structures that are, in profile, closely related. Each can be assessed individually, while considering them together provides additional information about the balance and continuity of the lower face — and the same principle applies to the relationships between temples, cheeks, nose and eyes.

Structure, proportion and harmony

Facial harmony is the impression that a face's features are proportioned consistently with each other and with the face's own scale — not the presence of any specific measurement. It is judged holistically, by eye, more reliably than it is captured by a single formula. Mathematical proportion can be a useful descriptive tool for discussing a face; the golden ratio specifically, as a claimed universal determinant of what makes a face beautiful, is a different and much stronger claim — and one the evidence does not support.

A 2024 review of the historical and clinical literature on the golden ratio in facial aesthetics concluded that there is currently no convincing evidence that the golden ratio is linked to idealised human proportions or facial beauty.

Facial thirds and fifths remain useful as reference points, not pass/fail criteria. They describe a common pattern, not a universal rule, and a face that departs from them is not thereby unbalanced — the departure is read against the rest of that individual face, not corrected toward an average.

Soft tissue changes the architecture

The skeleton sets the frame, but soft tissue is what is actually seen. Skin, fat and the connective tissue anchoring them to bone determine how a given skeletal structure reads — the same bony proportions can look rested or tired depending on the tissue lying over them.

Facial fat exists in distinct compartments, not one continuous layer, arranged in deep and superficial groups. The deep compartments, closest to bone, provide the underlying projection of cheek and midface; the superficial compartments contribute contour rather than structural support. CT-based study of the midface fat compartments in Asian women found that different compartments change in opposite directions with age within the same face — the infraorbital and nasolabial compartments tended to thicken, while the superficial cheek layer and part of the deep medial cheek fat tended to thin — which is why "volume loss" understates what is actually happening: some regions gain apparent fullness while adjacent regions hollow, changing the relationships between them rather than uniformly emptying.

Connective tissue is a further variable in this system. The retaining ligaments described by early anatomical studies anchor fat and skin to the underlying bone, but how much these specific attachments change with age, as opposed to the tissue around them, is not settled in the anatomical literature — what is well established is that bone, ligament, muscle and fat each age on their own timeline, at a pace that varies between individuals and between ethnic groups. The practical result is a face where contour can shift in one region independently of its neighbour, producing a face that looks hollow in one place and full in another at once. Assessing one area shows what is happening within it; assessing the surrounding region, or the face as a whole, can reveal transitions and relationships that may not be apparent from that area alone.

Movement and expression

A face cannot be fully understood from a single still image. Expression and the muscles producing it change how proportion, projection and symmetry read in real life, and a static photograph — however well lit — captures only one frame of that range.

The muscles of facial expression sit directly beneath the skin and attach to it in several places, which is why animation changes surface contour and not only apparent mood. A study comparing static photographs against dynamic video of the same smiles found that static images meaningfully underestimated features, such as gingival display, that only became visible in motion. Repeated contraction over years also has a lasting structural effect of its own, converting habitual expression lines into features visible even at rest.

An assessment made only at rest, from one angle, is an incomplete assessment. A face that reads as balanced in stillness can look markedly different mid-conversation, and a genuinely whole-face assessment considers both.

Masculine and feminine facial architecture

Male and female faces show consistent average differences in skeletal and soft-tissue structure — population-level tendencies, not a template either sex should conform to. On average, male faces tend to be larger with more central, forward projection from brow to chin, and female faces tend to be smaller with fuller cheeks and a more vertical forehead profile.

A 2026 three-dimensional geometric morphometric study of 342 healthy Italian adults aged 18 to 40 found that sex explained only around a tenth of the variation in facial shape within that group, and roughly a quarter of overall facial form once size was included — meaningful, but far from most of what makes one face look different from another. That specific figure describes one study population and should not be read as a fixed number for every population, including Asian ones, which have not been measured the same way; the more durable point is what the figure demonstrates in principle — that sex accounts for real but partial variation, leaving most of a face's individual character unexplained by sex alone.

Within Asian Facial Architecture™, these tendencies describe where a given face sits, not a target. A face with more traditionally masculine or feminine characteristics is not more correctly structured than one between them, and treatment aimed at moving a face toward either pole, rather than toward its own coherence, is optimising for the wrong thing.

How facial architecture changes with age

Facial ageing is not a single process. It involves bone, fat, muscle, connective tissue and skin, each on a different timeline, and describing it only as "volume loss" understates how much of what is seen is a change in the relationship between structures rather than a uniform loss of substance.

The facial skeleton remodels throughout adulthood. One recent literature review of facial bone ageing reported that, across the studies it examined, orbital enlargement averaged 15–20% and maxillary height decreased by 8–15% by the seventh decade, with the gonial (mandibular) angle typically becoming more obtuse over the same period — population averages drawn from the studies that review included, not a fixed timeline for any individual face, but indicative of how much the underlying scaffold can shift before most people notice a change on the surface. Facial fat compartments also change differently with age, with regional thinning and thickening altering contour rather than producing one uniform pattern of volume loss. Skin's own collagen content declines with age, but a meaningful proportion of visible skin laxity reflects the bone and fat changes beneath it rather than an independent skin process.

None of this is uniform across individuals or implies a fixed treatment response. It implies that an accurate assessment asks which structure has actually changed, rather than responding to a general impression of tiredness.

A feature does not exist in isolation

The face can be assessed at different levels: as individual features, as regions, and as a whole. Each perspective provides different information.

A chin, for example, can be assessed for its own projection, height and shape. But its aesthetic effect is also influenced by its relationship to the lips, nose, jawline and overall facial profile. Similarly, the appearance of the midface depends not only on cheek projection, but on the transitions between the lower eyelid, cheek, temple and lower face.

This does not mean that every feature must be treated as part of a whole-face intervention. A single, well-chosen change may be entirely appropriate. Nor does facial harmony require every feature to conform to a predetermined set of proportions.

Rather, it means that local assessment and whole-face assessment answer different questions. The first asks what is happening within a particular feature or region. The second asks how that feature participates in the face around it.

The decision may be local. The assessment should include context.

Within Asian Facial Architecture™, this distinction matters because the objective is not to optimise every feature independently. It is to understand which relationships are contributing to the appearance of the face, which are worth changing, and which should be left alone.

Assessment before intervention

Asian Facial Architecture™ treats assessment as a distinct step that precedes any decision about treatment. Its purpose is to sort a given face into three categories, not one:

• What may benefit from restoration — structure that has genuinely changed, in a direction most people would recognise as loss, and that treatment can plausibly address
• What may benefit from refinement — proportion or contour that has not changed with age but could be adjusted in keeping with the rest of the face, if wanted
• What should intentionally remain untouched — features or asymmetries that are part of that face's own baseline, where treating them as a "problem" would create disproportion rather than resolve it

Aesthetic medicine, approached this way, does not default to adding, filling or changing. A consultation that maps a face against these three categories, rather than against a single request, tends to end with a smaller, more precisely targeted plan. This is the same restraint-first approach that governs individual treatment decisions on this site, applied here at the level of the whole face.

A framework, not a formula

Asian Facial Architecture™ is not an attempt to define an ideal Asian face, and it does not reduce to a ratio, a set of measurements, or a template to move a face toward. It holds several things in view at once — skeletal structure, soft tissue, proportion, movement, expression, and the sequence in which a face ages.

What makes a given face work is rarely a single number. It is the relationship between its structures — and whether a proposed change preserves the coherence that makes that face recognisably its own. Assessing that accurately is the starting point for any recommendation, not an afterthought to one.