Image prompt
Generates a realistic variation of a reference bar photo by changing the subject's pose and interaction while preserving the original environment and candid style.
Copy directly into the target model and adjust for your input assets.
Using the provided reference image, create a realistic photo variation in the same dim Japanese bar setting, keeping the woman's outfit, hairstyle, stool, lighting, and candid smartphone-flash look consistent. Change her pose so she is turned more toward the man on the right and extends one arm forward, gently placing her hand on his shoulder as if getting his attention. Keep the man in the right foreground closer to the camera and partially blocking the view, with anonymized/blurred faces as in the reference. Preserve the cramped izakaya atmosphere, wall menus, bottles, dark wood, and casual candid composition.
Original source details are kept for attribution and review.
GitHub source：YouMind-OpenLab/gpt-image-2-prompts-search
This page is adapted from YouMind OpenLab GitHub prompt libraries. License: MIT。