Bc391f127009ec171109b379512d7fd5

POSITIVE PROMPT: Create a hyper-realistic split-scene bridal portrait that keeps the overall art direction, composition, atmosphere, and refined rendering style of the original gallery image, but visually divides the frame into two organically blended halves showing opposite realities of the same person. On the left half, depict the person from the uploaded photo exactly as in the previous description: an elegant bride standing confidently in a luxurious, bright art gallery, wearing a sophisticated white wedding dress with a long flowing skirt and subtle lace details, holding a large bouquet of vivid red roses in both arms at chest level, surrounded by high-end black-and-white framed portraits on pristine white walls and a polished reflective floor, with soft diffused gallery lighting and gentle spotlights creating clean highlights and soft shadows. On the right half, show the same person symmetrically positioned but transformed into a much more modest version: no makeup, hair unstyled and slightly messy, wearing a very simple, inexpensive light-colored dress, in a poor, worn environment that echoes the gallery layout but with rough, dull walls, cracked or stained flooring, and large portrait photos printed with low quality, faded tones and placed in precarious, mismatched frames. The division between the two halves should feel organic rather than a hard line, with textures, light and floor reflections subtly blending across the middle while still clearly contrasting luxury on the left and hardship on the right. Maintain realistic skin textures, consistent facial structure and body proportions across both halves so it is unmistakably the same person in two contexts. Keep lighting on the left as bright, soft, and flattering with high-key tones, while lighting on the right is dim, uneven and slightly cooler or more shadowed, with visible imperfections and insufficient illumination on the walls and portraits. Use an eye-level camera angle with a centered, full-body or three-quarter shot and a 50mm lens look, placing the person on both sides aligned along a central axis, mirroring pose but varying expression slightly to reflect confidence on the luxurious side and quiet resilience or fatigue on the poorer side. Explicitly ensure that the model must strictly preserve the identity of the person in the reference photo, keeping the same face structure, eyes, nose, mouth, jawline, proportions, hairstyle color, skin tone and gender with precise likeness in both halves, only adapting makeup, hair styling, clothing and environment to express the contrasting realities, and generate the image with no visible text, logos or signatures anywhere in the scene.
NEGATIVE PROMPT: deformed face, altered identity, extra limbs, duplicated heads, distorted anatomy, asymmetrical eyes, blurred face, wrong facial proportions, wrong jawline, wrong skin tone, wrong hair color, gender change, inconsistent facial features between halves, inconsistent hairstyle color, low resolution, artifacts, strong noise, over-smoothing, cartoonish style, extreme stylization, flat lighting, inconsistent perspective between halves, harsh straight split line, broken anatomy, bad hands, misplaced elements, any visible text, logos, watermarks, designer signatures, cropped face, distorted dress shapes, messy blending
DETAILS: [lighting: left side with soft diffused gallery lighting and subtle spotlights, bright high-key look; right side with dim, uneven, low-quality room lighting, visible shadows and low contrast, both sides blending organically at the center; camera: eye-level angle, full-body or three-quarter split-scene portrait; lens: 50mm lens look, consistent depth of field across both halves, keeping the subject and nearby elements sharp; composition: same person mirrored along the vertical center, left half in a luxurious bridal gallery with polished floor and high-end portraits, right half in a poor, worn room with simple low-quality photos and precarious frames, organic transition in textures and floor reflections at the middle; text: generate the image with no visible text, captions, logos or signatures anywhere in the scene]