GPT Image 2AI 提示词
Moody Portrait with Mosaic Face & Cozy Sweater — Sora 2 Prompt
Generate a hyper-realistic, DSLR-quality portrait of a woman in a ribbed turtleneck with an artistic warm-toned mosaic censoring her face, perfect for evocative imagery.
提示词 · EN
创作一张逼真的竖构图肖像，主体为 argsubject: 一位年轻女性，坐在室内窗边，光线柔和，取景范围从上半身到头部。她留着浓密的长波浪 arghair color: 深棕色头发，自然垂落在肩头，带有自然光泽和零星碎发。她穿着一件修身的罗纹针织 argsweater color: 深紫色 高领毛衣，长袖设计，强调织物纹理、纵向罗纹和自然的褶皱感。她的双手在画面下方的腰部轻轻交叠。面部被一个位于中心的大型不透明方形马赛克刻意遮挡，色调为暖棕色，遮住了眼睛、鼻子和大部分面部特征，仅露出部分头发和一只耳朵。可见的手部皮肤纹理需真实细腻，采用浅景深效果，左侧柔和的窗光，背景为暖色调的中性米色室内环境，呈现自然阴影，具有高细节的单反相机摄影质感。在左下角添加一个淡淡的半透明水印，内容为 argwatermark text: AI conia。
关于这个提示词
Generate a hyper-realistic, DSLR-quality portrait of a woman in a ribbed turtleneck with an artistic warm-toned mosaic censoring her face, perfect for evocative imagery. 它更适合作为 GPT Image 2 的概念艺术起点：先保留画面结构、主体关系和风格约束，再替换成你的品牌、人物或场景。
使用时优先替换 subject、hair color、sweater color、watermark text 等占位符，并保持镜头、构图和材质描述的顺序不变。这样更容易获得稳定的画面结果，也方便继续做多版本对比。
如何使用这个提示词
- Copy the entire prompt, including the variables in curly braces like
{subject}and{sweater color}. - Paste it into your chosen AI model's text-to-image or video prompt box (e.g., Sora 2, Midjourney, Stable Diffusion).
- Customize the default values inside the braces (e.g., change
深紫色to灰色for a gray sweater) to match your vision. - Generate the image, reviewing the output for lighting, texture detail, and the clarity of the mosaic effect.
- Adjust the prompt slightly if needed, perhaps by specifying camera lens type (e.g., '85mm portrait lens') for more control.
