PROMPT RADAR / PROMPT人像时尚

社交媒体帖子 - Cinematic Night Car Portrait

A photorealistic 85mm portrait prompt for a cinematic night scene inside a car, featuring dramatic streetlight shadows and professional color grading.

@AiwithElisia 0 热度 0 浏览
社交媒体帖子 - Cinematic Night Car PortraitORIGINAL VISUAL / 来源见文末
编辑先看

它为什么值得进入灵感库

这条内容值得留下的,不是表面的风格词,而是它提供了一条可以继续实验的画面路径。 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”。

完整视觉 Prompt

{argument name="subject" default="A cinematic, photorealistic close-up portrait of a young East Asian woman"} sitting in the back seat of a car at night. She has long, naturally straight, glossy black hair with a few loose strands falling softly across her face, delicate facial features, smooth natural skin texture, dark expressive eyes, subtle eyeliner, soft natural eyebrows, and muted rosy-red lips. She is looking directly toward the camera with a calm, neutral, slightly melancholic expression. {argument name="lighting" default="Warm golden horizontal bands of streetlight"} pass naturally across her face, creating dramatic cinematic light and shadow. The lighting should softly illuminate her eyes, nose, cheeks, and lips while leaving parts of her face and hair in deep natural shadow. Preserve realistic skin pores and fine facial details without excessive retouching or plastic-looking skin. She is seated inside a dark luxury car, with black leather seats and interior surrounding her. Through the rear window, a nighttime city scene is visible with heavily blurred colorful lights, warm yellow bokeh, white headlights, subtle red traffic lights, and dark urban buildings. Strong shallow depth of field separates her from the background. Shot on a professional full-frame camera, {argument name="lens" default="85mm"} portrait lens, f/1.4 aperture, cinematic depth of field, realistic low-light photography, soft filmic contrast, subtle warm highlights, deep blacks, natural color grading, high dynamic range, crisp eyes, detailed hair strands, realistic skin texture, soft bokeh, atmospheric nighttime mood. Composition: vertical portrait, subject centered slightly below the middle of the frame, head and shoulders visible, natural perspective, intimate close-up framing, cinematic negative space, realistic car interior framing in the foreground. Ultra-realistic photography, high detail, sharp facial focus, natural imperfections, realistic lighting, premium editorial photography, cinematic movie still, subtle film grain, professional color grading, 8K detail, no artificial beauty filter, no over-sharpening, no distorted facial features, no extra fingers, no text, no watermark.
查看原作者内容

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。

  1. 主体:A photorealistic 85mm portrait prompt for a cinematic night scene inside a car, featuring dramatic streetlight shadows and professional color grading.
  2. 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
  3. 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
  4. 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

保留作者,也保留判断。

本页收录公开案例并提供中文索引、结构化拆解和复刻建议,不冒充原作者作品,也不替代原帖上下文。

作者:Elisia · 收录于 2026年8月17日

访问原始来源

沿着这条线继续看