OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

复古海报墙旁的暖调人像

Warm-toned portrait beside a vintage poster wall

明确成年的女性身着得体的白色上衣与蓝色百褶裙，侧身倚靠在层叠复古海报墙旁；暖色柔光、细颗粒与低饱和色调构成自然怀旧肖像。

复古海报墙旁的暖调人像 — AI image creation
复古海报墙旁的暖调人像

提示词 / Prompt

中文提示词

创作一张自然写实的竖幅4:5复古编辑人像。主体为身份参考 ref-10a-v1 中明确成年的女性，留乌黑长发，穿着不透、领口得体的白色短袖衬衫，以及高腰蓝色百褶裙，裙摆过膝。服装保持不透且得体，裙长过膝。她略微侧身站立，轻松倚靠在旧墙旁，一只手自然扶墙，神情温和好奇，并带着轻松俏皮的感觉。采用平视或略低于眼睛的自然机位，构图尊重主体，不强调腿部或身体曲线。她身后的墙面层叠着原创的通用海报、传单和红、黄、蓝、白色图形。纸张呈现折痕、磨损和轻微翘边，但所有图案都保持通用、不可辨读；不要出现真实广告、品牌标志或可读文字。昏暗环境中的柔和暖光在人物与墙面形成细腻阴影。采用略微去饱和的复古色调、明显但细腻的胶片颗粒和轻柔光学扩散；主体面部与衣物清晰可辨，背景柔和虚化。保留自然肤质和真实衣料细节，营造怀旧而不挑逗的人像氛围。避免窥视式低角度、暴露短裙、可读的多语言文字、真实标志、水印、过度模糊和错误肢体。

English prompt

Create a natural, photorealistic vertical 4:5 vintage editorial portrait. The subject is the clearly adult woman in the configured identity reference ref-10a-v1, with long dark hair, wearing an opaque white short-sleeve blouse with a modest neckline and a high-waisted blue pleated skirt falling below the knee. Keep the clothing opaque and tasteful, with the skirt falling below the knee. She stands slightly turned and leans casually beside an aged wall, resting one hand naturally against it; her expression is gently curious with a relaxed playful note. Use an eye-level or only slightly low natural camera position and a respectful composition that does not emphasize legs or body shape. The wall behind her is layered with original generic posters, flyers and graphic shapes in red, yellow, blue and white. Show creases, wear and slightly lifted paper edges, but keep all artwork generic and unreadable, with no real advertisements, brand marks or legible text. Soft warm light in a dim setting creates delicate shadows on the subject and wall. Use a slightly desaturated vintage grade, noticeable but fine film grain and gentle optical diffusion; keep the subject's face and clothing clearly readable while softly easing the background. Preserve natural skin and fabric detail for a nostalgic but non-provocative portrait. Avoid voyeuristic low angles, revealing short skirts, readable multilingual text, real logos, watermark, excessive blur and malformed anatomy.

查看 API 模型

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作