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

东亚古风雪夜回眸电影感人像

Cinematic Portrait of a Glance Back on a Snowy Night in East Asian Classical Style

Poetic 2:3 winter-night identity portrait before a warmly lit traditional East Asian pavilion, with atmospheric snowfall and cinematic rim light.

东亚古风雪夜回眸电影感人像 — AI image creation
东亚古风雪夜回眸电影感人像

提示词 / Prompt

中文提示词

创作一张写实竖版2:3电影感冬夜人像：一位明确成年的主体置身于雪夜中的传统东亚场景。使用已配置的人物身份参考保留其身份、面部特征、肤色和可辨认发型，但重新设计服装、姿势、光线与背景。取膝部以上构图，采用优雅的四分之三背身姿势：身体略微转离镜头，头部越过肩膀回望观者，神情沉静、优雅并略带疏离。为其搭配精致的东亚传统风格冬装，采用层叠、不透视的面料，呈现丰富真实的织物细节，并加入适合其外观的雅致发饰；不要现代休闲装。背景是一座宏伟的多层传统东亚亭台或寺庙，暖金色灯光照亮建筑，建筑略微柔焦但结构合理。较密的自然雪花飘过画面，并在光线中清晰闪现。使用温暖的强逆光与克制的轮廓光勾勒头发和肩部，同时以柔和正面补光保持面部清楚自然。加入建筑灯光散景、空间大气透视、真实雪与衣料质感、自然比例，营造富有诗意、戏剧性且沉浸式的高端编辑人像氛围。面部准确对焦，曝光平衡。不要其他人物、扭曲解剖结构、畸形手部、重复五官、塑料皮肤、明显假雪、结构失真的建筑、文字、标志或水印。

English prompt

Create a photorealistic vertical 2:3 cinematic portrait of one clearly adult person outdoors at night in a snowy traditional East Asian setting. Preserve the person's identity, facial features, skin tone and recognizable hairstyle with the configured identity reference, but create a new outfit, pose, lighting and background. Show them from around the knees upward in an elegant three-quarter rear pose: body turned slightly away, head looking back over the shoulder toward the viewer, expression calm, poised and slightly distant. Dress them in refined traditional East Asian-inspired winter clothing with layered opaque fabrics, rich tactile detail and a tasteful hair accessory suited to their appearance; avoid modern casual styling. Behind them, a grand multi-story pavilion or temple is illuminated by warm golden lights, softly blurred but architecturally coherent. Heavy natural snowfall crosses the scene, with distinct flakes catching the light. Use strong warm backlight and a controlled rim light outlining hair and shoulders while a gentle frontal fill retains clear natural facial detail. Add soft bokeh from the architecture, atmospheric depth, realistic snow and fabric texture, natural proportions and a poetic, dramatic, immersive premium editorial mood. Keep the face in focus and exposure balanced. No extra people, distorted anatomy, deformed hands, duplicate features, plastic skin, artificial-looking snow, malformed architecture, text, logos, or watermark.

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