PROMPT RADAR / IMAGE插画三维

个人资料 / 头像 - Tokyo Train Schoolgirl Portrait

A photorealistic vertical portrait of a fashionable Japanese schoolgirl holding a train strap inside a Tokyo commuter train.

@mashiro_minase 0 热度 0 浏览
个人资料 / 头像 - Tokyo Train Schoolgirl PortraitORIGINAL VISUAL / 来源见文末
编辑先看

它为什么值得进入灵感库

它最有价值的部分,是把一个模糊想法推进到了可以被模型执行的程度。 轮廓语言、材质规则和世界观要互相支持,避免把互不相干的风格词堆在一起。

完整视觉 Prompt

Create a photorealistic vertical smartphone photo of a cute Japanese high school girl standing inside a modern Tokyo train during daytime. She is centered in a candid three-quarter pose, looking directly at the camera with a soft, slightly shy smile. She has long dark brown to black hair in twin ponytails, straight bangs, smooth fair skin, and natural makeup. She wears a cropped summer sailor school uniform: white short-sleeve top with a navy sailor collar and navy bow, showing a small amount of midriff, paired with a dark navy pleated skirt. Her right arm is raised above her head holding a white triangular train strap, and her left hand holds a smartphone with a multi-camera case. A large black shoulder bag hangs from her shoulder, decorated with exactly 12 visible plush keychains and charms: one large white fluffy plush, one brown teddy bear plush, one yellow doll-face plush, one pink plush character, one blue plush character, one white bear charm, one small beige animal charm, one round pastel charm, one tiny pink rabbit-like charm, one small multicolor capsule charm, one cream dangling plush, and one small red charm near the bag edge. The train interior includes metal poles and rails, windows with blurred city buildings outside, overhead Japanese advertisements, route signage, and soft natural daylight. Use realistic phone-camera perspective, slight wide-angle distortion, shallow depth of field, authentic candid composition, detailed fabric folds, reflective metal, and no illustration style. Keep the atmosphere youthful, fashionable, and everyday urban Japan. Optional customization: set the subject as {argument name="character name" default="a Japanese schoolgirl"}, hair as {argument name="hair color" default="dark brown-black"}, outfit as {argument name="school uniform style" default="cropped white-and-navy sailor uniform"}, location as {argument name="train location" default="Tokyo commuter train"}, and bag decoration theme as {argument name="plush charm theme" default="pastel cute character keychains"}.
查看原作者内容

不要一次改完所有变量

用 GPT Image 2 复刻时,先固定主体和构图,再逐次替换光线、色调或材质中的一个变量,这样更容易判断哪项描述真正起作用。

  1. 主体:A photorealistic vertical portrait of a fashionable Japanese schoolgirl holding a train strap inside a Tokyo commuter train.
  2. 视觉方向:社区实测摄影提示词;按原提示词的光线描述执行
  3. 镜头与构图:按原提示词的镜头与景别执行;按原提示词的构图与画幅执行
  4. 交付约束:auto,遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。
负向约束

遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。

模型建议

GPT Image 2 / Grok Imagine / Midjourney / Flux。先用原始 Prompt 建立基准结果,再根据模型对自然语言、镜头运动或文字排版的能力做局部改写。

来源与编辑原则

保留作者,也保留判断。

本页收录公开案例并提供中文索引、结构化拆解和复刻建议,不冒充原作者作品,也不替代原帖上下文。

作者:K's Al Photo Studio · 收录于 2026年8月14日

访问原始来源

沿着这条线继续看