Generate an ultra-photorealistic vertical selfie, 9:16. Use the TWO uploaded face references (a girl and a guy) and preserve identity strictly 1:1: 100% likeness, don't change features, proportions, age, eye/nose/lips/eyebrow shape, skin tone. Faces fully recognizable, no "beautifying."
Scene and pose: the girl and guy are sitting side by side on the edge of a metal grating platform at the very top of the Empire State Building in New York City, at an extreme height above Manhattan. Both looking at the camera, a light confident smile. Required: relaxed posture, weight shift, asymmetrical pose (asymmetry: shoulders at different heights, natural spine curve, torso slightly turned). Hands and fingers anatomically correct.
Camera/perspective: iPhone selfie aesthetic, wide angle ~24mm, slight perspective distortion like a front camera, camera slightly above eye level. Faces large in frame (head+shoulders), but the edge of the platform and the grating beneath them must be visible, and through the mesh — depth/emptiness and the Manhattan skyline far below, so the height reads immediately. No background blur — background should be legible.
Location (describe as precisely as possible):
— The uppermost service platform of the Empire State Building's spire/mast structure, a narrow area of galvanized steel open grating with square cells, visible metal ribs, edge profile trim, bolts/nuts/washers, mounting plates — consistent with the building's iconic Art Deco steel framework and antenna mast.
— Right next to the couple — a technical unit: a metal box with a control panel, several round buttons (red/black/metallic), switches/terminals, cables in conduit. No readable text/labels.
— Nearby, mounted aviation warning lights: two red dome lamps in metal housings (as used on the building's actual broadcast tower).
— Behind them — the building's lattice antenna/mast structure with steel diagonal braces, joint connections and cables, rising upward beyond the frame.
— Background: the recognizable Manhattan skyline far below — a dense grid of skyscrapers, the Hudson and East Rivers, Central Park in the distance, light atmospheric haze, clear sky and daylight (or soft golden hour), overwhelming sense of extreme height.
Outfits and accessories (strictly as in the references):
Girl:
— Clothing: a black T-shirt; Bottom — fully black, total-black aesthetic (black pants as in reference: heavy fabric/leather, high waist).
— Headwear: a black ribbed knit beanie with large "cat ear" details, as in reference 2. The beanie sits normally ON the head, NOT pulled down over the face; the face is fully open and clearly visible.
— Jewelry: large silver hoop earrings (several rings/cuffs in the ear as in reference 2) + a black choker snug around the neck.
— No masks/coverings on the face — face fully open.
Guy:
— Clothing: a black top tank , fitted, as in reference 1 (texture/ribbing visible, natural folds). Bottom — black (pants/trousers in a dark tone), overall strict total-black look.
— Accessories: no cap, no bandana/headband, no face mask — face and head fully open.
Quality/photorealism: photo raw style, Sony A7R look, ultra photorealistic, high detail, sharp focus on faces, natural colors, realistic shadows and contact shadow under the body/legs on the grating, no CGI.
Negative prompt (updated):
face change, identity drift, different makeup, different hair, any face covering on either person, balaclava/mask/bandana, cap/hat on the man, hat covering the woman's face/eyes, outfit change, wrong accessories, extra fingers/limbs, warped hands, floating, mismatched shadows, blur, plastic skin, CGI look, text, watermark, UI, logos.
