6706c3f0-a956-4930-aaf0-eddf1ba2b499

Two people sitting close together on the wooden steps of a weathered coastal cottage porch, late afternoon, overcast sky with diffused warm light breaking through low clouds near the horizon. Shot in 9:16 vertical. The dark-haired woman in her late twenties sits on the LEFT of the steps, legs drawn up, arms wrapped around her knees, her body angled slightly toward the man. She wears a chunky oatmeal wool sweater with sleeves pulled over her hands and dark jeans. Her dark hair is loose, wind-moved, a few strands across her face. She is looking at the man beside her, mid-sentence, her mouth slightly open, her grey-green eyes direct and unguarded for the first time — vulnerable, not performing. The man with brown hair and stubble in his early thirties sits on the RIGHT of the steps, one step lower than her, leaning back on his hands, looking up at her face. He wears a faded navy henley with sleeves pushed to his elbows, forearms weathered and tanned. He is listening — really listening — his expression open, warm, slightly pained, like someone hearing a beautiful thing they know will hurt. Between them: a gap of about six inches, charged, neither touching. Two mugs of tea on the step between them, steam still rising faintly. His dog — a medium brown mutt — lies on the porch behind them, chin on paws, asleep. COMPOSITION: she is upper-left of the vertical frame on the higher step, he is lower-right on the lower step — the height difference puts her eyes above his, reversing the usual dynamic. The porch railing frames them on the right. Behind them through the railing gaps: the grey Atlantic, the boatyard, fog sitting low on the water. Below them: the gravel path, wet. LIGHTING: no artificial light. Late afternoon overcast with a single break in the clouds near the horizon casting warm diffused amber light from the left side, hitting her face and his forearms. The rest of the scene is cool blue-grey overcast shadow. The contrast between the warm break-light and the cool ambient is the entire palette — amber faces against a grey world. No reflectors, no fill, no bounce. The light falls where physics puts it. Shadows under the porch roof are deep and real. LENS AND TEXTURE: shot on full-frame mirrorless, 40mm at f2.2 — intimate but not claustrophobic. Shallow enough that the ocean behind the railing goes soft but both faces stay sharp. Hyperrealistic at the highest level — every skin pore visible on both faces, individual stubble hairs on his jaw, the wool weave of her sweater, the grain of the weathered porch wood, the steam rising from the mugs, salt residue on the railing, her flyaway hairs lit by the amber break-light like fine copper wire. His forearm veins. Her chapped lower lip. The dog's fur texture. This should look like a photograph taken by a documentary photographer who happened to witness an intimate moment — not staged, not lit, not art-directed. Found beauty. 9:16 vertical. negative prompt cinematic lighting, color grade, teal-orange, studio light, fill light, bounce card, reflector, golden hour perfection, beauty filter, smooth skin, airbrushed, glamour, makeup, styled hair, clean new clothes, new porch, CGI, 3D render, anime, illustration, lens flare, bokeh balls, horizontal framing, posed model sitting, looking at camera, smiling, happy, indoor, night, artificial light, perfect symmetry, stock photo

Generate a single, hyper-realistic character reference image (strictly no collages, text, watermarks, or borders). The image must feature a cinematic, three-quarter portrait of the described character, maintaining a sharp, tack-focused emphasis on the face with natural skin textures, pore details, and realistic micro-expressions. The lighting should be cinematic and directional (e.g., subtle split or Rembrandt lighting) to give depth, blending seamlessly with the environment specified in the character description. The character must have a natural, living presence, looking slightly off-camera or directly at the lens with a subtle, context-aware expression. Ensure the full head, neck, and upper shoulders are completely unobstructed. This image serves as a high-fidelity identity blueprint for temporal-consistent video generation.