PROMPT RADAR / PROMPT人像时尚
个人资料 / 头像 - Golden Hour Outdoor Portrait
A prompt for a peaceful, ultra-realistic golden hour portrait of a young woman with her eyes closed.
@Sairah_0 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
Ultra-realistic golden hour close-up portrait of a young woman outdoors, eyes gently closed with a soft peaceful smile, face tilted slightly upward toward the warm setting sun, {argument name="hair style" default="loose windblown dark brown hair flowing naturally across her face"}, glowing sun-kissed skin with natural texture and subtle freckles, minimal makeup, soft glossy lips, wearing a {argument name="clothing" default="loose white linen blouse with a delicate collar"}, cinematic warm amber sunlight creating beautiful highlights and gentle shadows, shallow depth of field, clean blue sky background, dreamy summer atmosphere, authentic candid expression, ultra-detailed skin texture, high dynamic range, 85mm portrait lens, f/1.8, professional fashion photography, natural color grading, soft contrast, editorial lifestyle aesthetic, photorealistic, 8K, RAW quality.
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 a peaceful, ultra-realistic golden hour portrait of a young woman with her eyes closed.
- 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