GPT Image 2AI 提示词
Censored Track Athlete Portrait — Sora 2 Prompt
Generate a photorealistic vertical portrait of an adult female track athlete in a purple racing uniform against a concrete wall for sports-body-type comparison visuals.

提示词 · EN
Create a realistic vertical smartphone-style portrait photo of an adult female track athlete standing front-facing against a plain gray concrete wall with visible panel seams, small circular form-tie holes, and subtle weathered texture. She has a soft, curvy athletic build and fair skin, with her arms held behind her back and shoulders relaxed, cropped from the top of the head to mid-thigh. Her hair is arghair color: muted purple, tied back loosely with a few strands around the ears. Her face is intentionally covered by a centered opaque rectangular censor block in a skin-toned mauve color, hiding all facial features. She wears a glossy arguniform color: purple two-piece track-and-field racing uniform: a racerback sports crop top and matching briefs, both with metallic silver-gray side panels, thin white curved piping, and small white swoosh-like athletic logos on the chest and hip. The pose is neutral and documentary-like, evenly lit by soft outdoor daylight, with realistic skin texture, fabric wrinkles, slight shadows under the arms and torso, and natural perspective. Use a clean candid sports portrait aesthetic, high-resolution photorealism, shallow but not blurry depth of field, no text, no extra people, no props.
关于这个提示词
Generate a photorealistic vertical portrait of an adult female track athlete in a purple racing uniform against a concrete wall for sports-body-type comparison visuals. 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 hair color、uniform color 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 full prompt text from the page.
- Paste it into your preferred AI image generator (like Sora, Midjourney, or Stable Diffusion).
- Customize the variables in curly braces (e.g., change hair or uniform color) to match your vision.
- Set your desired output aspect ratio to vertical/portrait orientation.
- Generate the image and refine as needed with additional tweaks.



