PROMPT RADAR / IMAGE插画三维
个人资料 / 头像 - Casual Artoria Anime Portrait
A refined anime portrait of a blonde heroine in a cream off-shoulder sweater and jeans, suited for character fan art or casual fashion illustration.
@uniuni_main_acc 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。 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。
01 / 原始方法
完整视觉 Prompt
Create a polished anime-style half-body portrait of {argument name="character name" default="Artoria Pendragon"} as a casually dressed young woman with fair skin, bright teal-blue eyes, and {argument name="hair color" default="pale blonde"} hair. She has straight bangs framing her face, one long ahoge strand curving upward from the crown, and her hair pulled back into a neat braided bun secured with a large dark navy ribbon bow visible behind her head. Show her from about mid-thigh upward, standing at a slight angle with a gentle confident smile, head subtly tilted, one hand resting on her hip and the other arm relaxed downward. Her outfit contains exactly 2 visible clothing pieces: a cream-white ribbed off-shoulder knit sweater with long sleeves pushed slightly up at the forearms, and high-waisted dark blue denim jeans with belt loops, metal button, stitching, and front pockets. Emphasize soft shoulders and collarbones, delicate facial features, clean line art, smooth cel shading, and refined anime illustration detail. Use warm indoor studio lighting with golden bokeh highlights, a softly blurred neutral gray-beige background, and a bright lens flare glow in the upper right. The composition should be vertical, intimate, elegant, and softly romantic, with no text, no watermark, and no extra characters.
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 refined anime portrait of a blonde heroine in a cream off-shoulder sweater and jeans, suited for character fan art or casual fashion illustration.
- 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