Prompt 1
A scene featuring the person in the uploaded reference image (DO NOT TOUCH FACE). Image 1 is the identity-neutral chroma-green scene template. Image 2 is the target identity reference. Scene: A person is crouched on a sandy beach, facing away from the camera toward the ocean. The lighting is natural daylight under a partly cloudy sky. Wardrobe: The subject is partially nude, wearing a brown short-sleeved t-shirt, a green baseball cap, and a patterned thong bikini bottom. Face: The subject's face is completely obscured from view as they are facing away from the camera. Wardrobe geometry authority: Image 1 controls the exact neckline depth and shape, strap placement, visible cleavage and coverage, bust contour created by the garment, hem, fit, fabric tension, and overlap. Do not raise, flatten, widen, or redesign the garment. Wardrobe appearance authority: preserve the garment colors, patterns, sheen, opacity, and fabric texture visible in Image 1; never treat clothing as mannequin material. Hair authority: Image 1 controls hairstyle geometry and style, including length, silhouette, part line, root direction, volume, layered sections, bangs, braids or tied sections, wave or curl groups, separated locks, overlaps, tapered ends, and whether the hair is worn up or down. Image 2 controls the target subject's hair color and identity. Interpret every chroma-green hair section as a geometry guide, not as a solid object. Reconstruct complete photorealistic natural hair with continuous strands and fine texture while preserving the guided section placement and flow. Never return the target subject bald, and never leave a helmet, cap, slab, wedge, faceted polygon shell, or blocky low-poly hair shape. Replace only the primary mannequin with Mila from Image 2. Replace every secondary mannequin with a separate, distinct new adult identity that roughly preserves that secondary subject's visible hairstyle silhouette, styling, clothing, pose, scale, and scene role from Image 1. Use Image 2 exclusively for the primary subject. Never copy or duplicate Mila onto a secondary subject. Do not create twins or clones. Remove all chroma-green mannequin material, green spill, synthetic texture, and placeholder traces from every subject. Preserve legitimate green scene objects and clothing. Match the stated action, camera relationship, face, hand positions and interactions, visible object inventory, person count, pose, crop, clothing, lighting, background, and composition. Do not add an implied phone, camera, mirror, held prop, person, or text that the Scene does not show. Do not preserve any original source identity. Photorealistic adults, natural skin texture, anatomically coherent faces, hands, and bodies. Critical user correction: Redo the mannequin template to preserve detailed natural hair form: keep separated layers, locks, wave or curl groups, overlaps, and tapered ends; avoid helmet, slab, faceted, polygonal, or blocky hair. Use Seedream V5 Pro for the final replacement.
- Engine
- Seedream V5 Pro
- Dimension
- Portrait (3:4)
- Gender
- Female
- Ethnicity
- White
- Age
- Twenties









