PROMPT RADAR / PROMPT人像时尚
个人资料 / 头像 - Hyperrealistic Mirror Selfie Prompt
A prompt for creating a realistic iPhone mirror selfie in an elevator, preserving subject identity from a reference.
@MissDelulu9 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
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。 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
Preserve the clearly adult woman’s exact identity, facial features, proportions, skin tone, and natural asymmetry. Hyperrealistic Pinterest-style iPhone 15 Pro Max mirror selfie inside a {argument name="setting" default="modern brushed-steel elevator"}. She leans lightly against the wall, one leg straight and the other bent with foot resting naturally against the wall, relaxed elegant pose. She holds an iPhone beside her cheek while her other hand carries {argument name="accessories" default="pink roses and a white shopping bag"}. {argument name="attire" default="Fitted black mini dress, oversized black blazer, black strappy heels"}, delicate necklace, small hoop earrings, light makeup. Dim overhead lighting, cool metallic reflections, realistic skin texture, authentic phone-camera grain, natural mirror perspective, full-length vertical 9:16, sharp face and clothing, believable hands and reflections. --no altered identity, extra fingers, extra people, duplicated reflection, warped phone, CGI, plastic skin.
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：A prompt for creating a realistic iPhone mirror selfie in an elevator, preserving subject identity from a reference.
- 视觉方向：社区实测视觉提示词；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