PROMPT RADAR / PROMPT人像时尚
社交媒体帖子 - Cheerful Spring Park Pinafore Girl
A vibrant outdoor spring portrait prompt of a young girl in a park, focusing on joyful expression and traditional fashion elements like Mary Janes and a pinafore dress.
@woleswoosh 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
Funny Woman
A cheerful young East Asian girl in her mid-teens standing outdoors on a sunny spring day, full-body shot from a slight low angle. She has fair skin, soft black hair styled in two long twin braids with neat bangs, and is wearing thin round wire-rimmed glasses. She smiles brightly with her mouth open, looking directly at the camera with a warm, joyful expression.
She wears a short-sleeved cream-colored blouse with delicate small {argument name="embroidery" default="floral embroidery"} in soft pink and green tones, layered under a bright {argument name="dress color" default="orange"} sleeveless pinafore dress with a fitted bodice, thin straps, and a flared knee-length skirt that catches the light. In both hands she holds a yellow-and-white checkerboard pattern canvas tote bag in front of her waist. On her feet are white ankle socks and classic black Mary Jane shoes with a single strap.
The background is a lush public park filled with vibrant green trees and dense foliage under a clear bright blue sky. Orange {argument name="flower type" default="cosmos flowers"} and other wildflowers bloom in the grass around a low wooden rail fence. Soft natural sunlight creates gentle shadows and a warm, airy atmosphere. Photorealistic style, high detail, natural lighting, shallow depth of field, vibrant yet soft spring color palette.
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 vibrant outdoor spring portrait prompt of a young girl in a park, focusing on joyful expression and traditional fashion elements like Mary Janes and a pinafore dress.
- 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