Prompting Portraits and Headshots

Portrait photographers use a vocabulary of named lighting patterns that AI models have learned from decades of photographic training data. Using these names in your prompts is the most efficient way to specify the three-dimensional light arrangement: where the key light falls, what the shadows do, and how the face is shaped by light. Rembrandt lighting: 'Rembrandt portrait lighting, key light at 45 degrees to subject's left and slightly above eye level, characteristic small triangle of light on the shadowed cheek, deep shadow on the right side of the face, warm single source.' Three-point lighting standard: 'three-point portrait lighting, key light camera-right, fill light camera-left reducing shadow by one stop, hair light from upper rear separating subject from background.' Butterfly lighting (glamour): 'butterfly lighting portrait, key light directly above and slightly forward of the face, characteristic butterfly shadow under the nose, flattering for high cheekbones, clean and glamorous.' Split lighting: 'split lighting portrait, key light at 90 degrees to the face, exactly half the face in deep shadow, graphic and dramatic, suitable for character or editorial portraits.' Broad light: 'broad lighting portrait, key light on the side of the face turned toward camera, fuller-faced appearance, friendly and approachable quality.' Short light: 'short lighting portrait, key light on the side turned away from camera, face appears more dimensional and slimmer.' Using the pattern name and then specifying the light quality — 'large soft-box key light' vs. 'bare hard light' — gives the model complete lighting information in under twenty words.

Expression direction is where portrait photography converges with actor direction. AI models respond better to physiological description of expression than to emotion labels, because emotion labels are ambiguous. 'Confident' could mean fifty different facial configurations; 'direct gaze to camera, slight upward tilt to chin, relaxed jaw, neutral mouth with just the faintest upturn at the corners, eyes fully open and engaged' produces a specific and consistent result. For professional headshots, the standard expression variables are: gaze (direct to camera or slightly off-camera at 15 to 30 degrees); mouth position (closed, slight smile, full smile with teeth, relaxed open); jaw tension (relaxed vs. defined); and eye engagement (fully open, slightly narrowed in concentration, crinkled in genuine warmth). For corporate professional headshots: 'direct confident gaze to camera, open and engaged expression, slight natural smile, relaxed jaw, professional composure.' For creative or personality-driven headshots: 'three-quarter gaze to upper left, slight knowing smile, one eyebrow marginally raised, relaxed and self-assured quality.' For acting headshots: 'subtle emotional presence, slight vulnerability in the eyes, mouth soft and neutral, available and present quality, the subject is listening rather than performing.' The gaze direction creates very different psychological effects: direct gaze engages the viewer in a confident relationship; off-camera gaze suggests the subject has an inner life and the viewer is observing a real moment. Choose based on the intended use case and relationship the headshot needs to establish with its audience.

Skin rendering quality is the most technically demanding aspect of portrait generation. Default AI skin often falls into one of two failure modes: over-smoothed and plastic-looking (the model has averaged out real skin texture into a perfect but inhuman surface) or over-detailed and papery (the model has overcompensated toward texture and created something that looks like dried leather). Realistic skin has specific visual characteristics that must be explicitly requested. Skin surface detail: 'visible pore texture in the T-zone and cheek area, slight texture variation between oily and dry zones, no visible airbrushing or smoothing, natural subsurface scattering giving skin a translucent quality.' Subsurface scattering is the technical phenomenon that makes skin look luminous rather than opaque — light enters slightly below the skin surface and scatters back out, creating the characteristic warm glow. Prompting for it directly: 'subsurface light scattering in the skin, translucent warm quality especially at the ears and lips, light visible through the skin at thin areas.' Skin tone rendering accuracy: 'accurate and even skin tone rendering, no color shifts between shadowed and highlight zones, correct hue maintained throughout.' For darker skin tones, which models historically render less accurately: 'rich deep brown skin tone, accurate color in both highlights and shadows, no desaturation or graying of the dark tones, warm undertone maintained, natural subsurface glow.' Specifying 'photographic realism, skin texture indistinguishable from professional portrait photography' sets the quality bar for the model's rendering decisions.

The background in a portrait is not neutral space — it carries significant meaning about context, professionalism level, and personal brand. The wrong background undermines an otherwise excellent portrait. The most common portrait background types and their prompt patterns: Simple studio backdrop: 'plain mid-grey seamless paper backdrop, even illumination, no texture, professional studio portrait look.' Gradient studio: 'gradient background smoothly transitioning from dark charcoal behind the subject to lighter grey to the sides, subtle vignette at corners, depth created by the gradient.' Environmental: 'out-of-focus office environment visible in background, warm bokeh from office lights, glass walls suggesting a modern tech company, subject clearly separated from background.' Outdoor environmental: 'shallow depth of field, outdoor city environment behind, buildings and pedestrians blurred to smooth bokeh, depth of field clearly separating subject from background.' Abstract bokeh: 'circular bokeh background, large aperture lens quality, out-of-focus colored lights forming abstract warm background, studio quality light on the face.' For corporate headshots, the background needs to feel professional without distracting from the subject: 'medium-toned grey background, subject clearly separated with slight edge glow, professional corporate headshot background.' Subject-to-background separation is achieved through lighting and depth of field: 'hair light creating bright rim light separating dark hair from dark background' or 'wide-aperture lens blur isolating subject from background environment.' Without explicit separation instruction, models often merge the subject into the background at the edges.

Clothing communicates professional context and personal brand as powerfully as any other visual element in a headshot. A great portrait can be undermined by clothing that signals the wrong professional register. Specify clothing with the same care you apply to lighting: garment type, color, fabric, fit, and styling details. For corporate professional: 'dark navy structured blazer, crisp white dress shirt, no tie, neat collar, tailored fit, clothing communicates executive professionalism.' For creative professional: 'well-cut dark turtleneck, minimal jewelry, deliberate and considered styling, confident creative professional register.' For tech startup or casual professional: 'clean white or grey crew-neck t-shirt, good fit, casual but intentional, approachable and modern.' For medical or academic professional: 'white lab coat over neat professional clothing, name badge visible at lapel, scientific professional context.' Always specify collar and neckline framing because the transition from clothing to neck to face is a critical compositional zone in a portrait crop. 'V-neck collar opening toward camera, slight collar gap showing collarbone' reads very differently from 'closed high collar, structured and formal.' Hair and grooming specificity matters equally: 'hair neatly styled but not rigidly so, slight natural movement, professional grooming level, not overly formal.' These styling details shift the subject from generic AI person to a portrait that feels like a deliberate, consistent personal brand.

Different headshot applications have different conventions that audiences have absorbed unconsciously. Meeting those conventions signals competence and belonging in the context; violating them creates friction even when viewers cannot articulate why. LinkedIn and corporate directory headshots: clean background, direct gaze, genuine but restrained smile, professional clothing, approximately face-to-shoulders crop. 'LinkedIn professional headshot, clean medium-grey backdrop, direct confident gaze to camera, slight natural smile, dark blazer, face and shoulders framing, sharp focus, even professional lighting.' Acting headshots: slightly wider crop showing more of the body, expression that conveys presence and availability rather than fixed professional smile, off-camera gaze often preferred, environment or natural light frequently used. 'Actor headshot, slightly warm natural lighting, three-quarter crop to mid-chest, slight off-camera gaze, emotionally available expression, natural and real quality.' Personal brand and entrepreneur headshots: the photograph should communicate the brand personality explicitly — choose warmth or authority, approachable or expert, creative or corporate and then build every element of the portrait around that positioning. 'Personal brand headshot for creative consultant, warm environmental background suggesting a creative workspace, genuine and warm expression, smart-casual clothing, energy of an approachable expert.' Real estate agent headshots have their own conventions: 'real estate professional headshot, confident friendly expression, well-tailored business attire, clean background, trustworthy and local-expert quality.' Knowing the convention and either meeting it precisely or deliberately subverting it gives you control over the audience response.