
原始文档
GPT Image 2 · en
Scene:
A square underground subway platform rendered as a polished editorial image, viewed along the platform edge with strong perspective lines, glossy floor reflections, cool fluorescent light, and a train window acting as a long horizontal mirror. The space is clean and cinematic, with no readable signage or route information.
Subject:
Show the same woman at five life stages distributed along the platform: small child holding a folded coat, teenager listening quietly, young adult with a travel bag, middle-aged adult looking into the train window, and elderly version reflected inside the dark glass. Preserve identity through the same high cheekbones, dark arched brows, short chin, and mustard yellow coat motif adapted by age. Each version should appear at a different depth point while still reading as one person across time.
Important details:
The subway platform must behave like one continuous architectural space. Reflections in the train window and floor should be plausible, with no duplicate bodies in impossible places. The time-split should be obvious through pose, scale, age, and costume changes, not through text. Keep facial continuity stronger than clothing differences.
Use case:
Mirror Worlds single study 6/8, testing life-stage identity consistency along a reflective subway platform without references.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no crowded commuters, no route numbers, no advertising posters, no inconsistent face, no train-brand marks, no warped platform geometry.使用方法
- 复制完整原始提示词,并保留其中的参数。
- 按需求替换主体、文字和构图要求。
- 在兼容的外部工具中使用,核对当前使用条件并检查结果。
开始前需了解
- 先确认是否需要参考图。标注兼容模型,不代表该示例由这个模型生成。
- 核对名称、数字与画面文字;不同模型版本的结果可能不同。





