Gpt-Image-2 On Chatgpt Prompt: Using Image A (Face)

the prompt

Using Image A (face) and Image B (full body), create a photorealistic vertical snapshot of the same young adult woman in her early twenties, preserving her identity and proportions.

She is sitting on a concrete stairwell, shot from a low angle near her feet. The composition is shoe-dominant: her powder-pink satin low-profile sneakers inspired by Miu Miu Plume are closest to the lens and sharpest in focus. Crop the face around nose level, keeping her mouth, jawline, hair, and outline visible enough to feel like the same person. She leans slightly forward with her arms resting on her knees or thighs, calm and unreadable.

She wears white rib socks, washed black denim bermuda shorts, and a faded gray crewneck sweatshirt.

Make it look like a rough Utsurundesu flash snapshot: harsh direct flash, strong flash falloff, dark background, heavy film grain, slight misfocus, slight motion blur, slight camera shake, uneven exposure, muted washed colors, imperfect framing, gritty failed-snapshot texture.

Avoid ballet-flat shapes, mary-jane straps, polished commercial photography, hyper-clean sharpness, plastic skin, overly cute styling, anime style, distorted anatomy, underage appearance, text, logos, and watermarks.