OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

雪の夜、暖かな灯りの東アジア風亭台の人物像

Portrait by a Warmly Lit East Asian Pavilion on a Snowy Night

「雪の夜、暖かな灯りの東アジア風亭台の人物像」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

雪の夜、暖かな灯りの東アジア風亭台の人物像 — AI image creation
雪の夜、暖かな灯りの東アジア風亭台の人物像

プロンプト

中国語プロンプト

创作一张写实、电影感的竖版雪夜人像,主体为一位明确成年的人物,面部身份与具有辨识度的发型绑定至已配置的ref-10a-v1参考。保持该身份,不要替换成另一张脸,也不要臆造身份特征。画面从大腿中部向上取景,人物以优雅的三分之二侧身姿态站立,身体略微转离镜头,再回头越过肩膀望向观者,神情沉静从容。为人物穿着优雅的东亚传统风格冬装,采用不透明的叠层面料、精致刺绣与克制的发饰,保持面部无遮挡;不要声称属于某个具体朝代或历史服制。夜晚户外背景设置一座宏伟的多层东亚亭台建筑,层叠飞檐由温暖金色灯光照亮,并在远处柔和虚化。让充足且清晰可见的雪花穿过画面并捕捉灯光,同时保证雪不会遮挡面部。以受控的暖色轮廓光勾勒头发与肩部,与冷蓝色夜间环境光形成对比,辅以柔和散景和浅景深;保留真实皮肤与织物质感以及自然协调的人体比例。人物、亭台和飘雪应有清晰可读的空间层次,营造诗意而有戏剧感的冬日编辑摄影氛围。若来源图片实际可用,仅将其作为构图/风格参考,不作为身份替代。不要文字、标志、水印、重复人物、身份改变、畸形手部、现代休闲服或不自然的假雪。

英語プロンプト

Create a realistic vertical cinematic night portrait of one clearly adult person, with facial identity and recognizable hairstyle bound to the configured ref-10a-v1 reference. Preserve that identity; do not substitute another face or invent identity-defining traits. Use the listed source image only as a planned composition/style reference if it is actually available, never as a replacement identity reference. Show the subject from mid-thigh upward in a graceful three-quarter stance, body angled slightly away and head turned back over the shoulder toward the viewer, with a calm, composed expression. Dress them in elegant traditional East Asian-inspired winter clothing with layered opaque fabrics, refined embroidery and tasteful hair accessories that do not obscure the face; avoid claiming a specific dynasty or historical garment. Place them outdoors at night before a grand multi-story East Asian pavilion with layered eaves, illuminated by warm golden architectural lights and softly defocused in the distance. Let abundant distinct snowflakes fall through the scene and catch the light, while keeping the face clear and unobscured. Use a controlled warm rim light around hair and shoulders against cool blue night ambience, gentle bokeh and shallow depth of field; retain realistic skin and fabric texture and coherent natural proportions. Keep the person, pavilion and snowfall separated into readable depth layers for a poetic, dramatic winter-editorial mood. No text, logo, watermark, duplicate subject, altered identity, malformed hands, modern casual clothing or artificial-looking snow.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成