PROMPT RADAR / IMAGE人像时尚
个人资料 / 头像 - Adjust Glasses Placement in Bed Selfie
Creates a minimal image edit that only repositions eyeglasses on a photorealistic reference selfie while preserving identity and scene details.
@takada_merca 0 热度 0 浏览
ORIGINAL VISUAL / 来源见文末编辑先看
它为什么值得进入灵感库
先别急着照抄整段 Prompt:真正有用的是主体、空间和镜头之间的约束关系。 先锁定人物身份与服装轮廓,再用光线、焦段和动作消除“普通写真感”。
01 / 原始方法
完整视觉 Prompt
Edit the provided reference image and change only the glasses placement: make the round metal glasses sit more naturally and slightly lower/centered on her face while still covering the eye area. Preserve the same clearly adult Japanese woman, recognizable identity, black bob with bangs, lying-in-bed selfie pose, close front-camera angle, sleepy wink expression, duvet covering her mouth and nose, visible arm, bedding, hotel-room background, lighting, camera perspective, and overall photorealistic smartphone snapshot look. Do not change her face, age, hairstyle, pose, clothing visibility, environment, or add any new objects.
02 / 复刻路径
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。
- 主体:Creates a minimal image edit that only repositions eyeglasses on a photorealistic reference selfie while preserving identity and scene details.
- 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
- 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
- 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