OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image

Korean Photo Booth Six-Frame Character Reference Collage

Korean Photo Booth Six-Frame Character Reference Collage — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.

Korean Photo Booth Six-Frame Character Reference Collage — AI image creation
Korean Photo Booth Six-Frame Character Reference Collage

Prompt

Chinese prompt

创作一张超写实 PHOTOISM 风格的韩国照片亭拼贴,采用精确的 3×2 网格:三列两行,共 6 张彼此区分的照片。仅使用已绑定到 ref-10a-v1 的 identity_reference 作为唯一人物参考,并确保六帧中始终保留同一张真实且可辨识的脸。每帧随机选择不同服装,同时保持人物身份和照片亭环境一致;不需要另行提供外观参考图片。每帧均加粗黑色 PHOTOISM 风格边框,并在每帧右侧边缘清晰竖排准确文字“MY DAY”。场景设于室内照片亭,背景为纯浅灰墙面;使用温暖的 iPhone 风格直闪,略微过曝,并加入柔和演播室补光和真实照片亭胶片颗粒。姿势与表情俏皮、荒诞且充满活力,同时保持面部特征清晰一致和高分辨率写实细节。网格干净利落、焦点清晰、对比度较高,呈现精致的照片亭美学。除每帧重复竖排的“MY DAY”外不得出现其他文字;避免色彩过于鲜艳、背景杂乱、模糊、低质量、面部畸变、手部畸形、比例错误、排版混乱、噪点过多、卡通风格、不真实细节、标志或水印。

English prompt

Create one hyper-realistic PHOTOISM-style Korean photo-booth collage as an exact 3×2 grid: three columns and two rows, containing exactly 6 distinct photographs. Use the attached identity_reference bound to ref-10a-v1 as the sole character/person reference and preserve the same recognizable real face consistently in all six frames. Give each frame a different randomly chosen outfit while keeping the person and photobooth setting coherent; do not require a separate appearance-reference image. Add a thick black PHOTOISM-style border to every frame, with the exact text “MY DAY” clearly set vertically along the right edge inside each frame. Use an indoor photo-booth setting with a solid light-gray wall, warm iPhone-style direct flash with slight overexposure, soft studio fill and authentic photobooth film-grain texture. Make the poses and expressions playful, absurd and energetic while retaining clear, consistent facial features and realistic high-resolution detail. Keep the grid clean, sharp and high-contrast with a sophisticated photobooth aesthetic. Do not add any text beyond the repeated vertical “MY DAY”; avoid overly vibrant colors, a messy background, blur, low quality, distorted faces, malformed hands, bad proportions, messy typography, excessive noise, cartoon style, unrealistic details, logos or watermarks.

View API model

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas