OpenZhenZhen AI 提示词画廊 · GPT Image 2.6 · image

都市河畔沉思者电影感全身肖像

Cinematic Full-Body Portrait of a Thinker by the Urban Riverside

将来源成年男性人物适配为 ref-10a-v1 中的成年女性，在都市河畔按来源指定坐姿、服饰、配饰与阴天侧光完成全身环境肖像。

都市河畔沉思者电影感全身肖像 — AI image creation
都市河畔沉思者电影感全身肖像

提示词 / Prompt

中文提示词

将 identity_reference ref-10a-v1 作为唯一面部身份参考。把来源中的二十多岁成年男性适配为实际参考中的同一位成年女性；不要声称她是男性，也不要给她指定缺乏依据的精确年龄。创作一张照片级写实的电影感全身生活肖像：她坐在现代都市滨河步道旁的低矮混凝土台沿上。她神情平静、若有所思，目光略带疏离并望向镜头旁。一侧膝盖弯起，另一条腿放松垂下；一只手肘靠在抬起的膝盖上，手指轻触太阳穴。她戴一顶无字的纯色中性棒球帽，穿结构化深橄榄色夹克，内搭浅米色 T 恤，配炭灰色修身休闲裤、米白色极简皮革运动鞋、细链项链、一枚简约戒指和薄金属框圆眼镜。项链是来源文字指定的造型配饰，并非从面部参考图提取。画面展现带微妙反光的光滑石板路、沿河金属栏杆、柔焦城市建筑和阴天漫射天空。采用柔和自然侧光和轻柔面部阴影，营造忧郁但干净的杂志氛围，使用柔和大地色调与低反差，保持人物清晰、皮肤纹理自然。以 50mm 定焦、f/2.8、平视略侧的摄影观感作为视觉目标，不声称实际使用了特定相机或设置。全身与鞋子均须入镜，同时保留可辨认的环境信息。避免身份漂移、文字、水印、额外人物、解剖畸形、强烈反差和过度虚化的环境。

English prompt

Use identity_reference ref-10a-v1 as the sole face-identity reference. Adapt the source's adult male in his mid-twenties to the same adult woman shown in the actual reference; do not claim she is male or assign her an unsupported exact age. Create a full-body photorealistic cinematic lifestyle portrait of her seated on a low concrete ledge beside a modern urban riverside promenade. Her expression is calm and thoughtful, with a slightly distant gaze just off-camera. One knee is bent upward while the other leg hangs relaxed downward; rest one elbow on the raised knee and let the fingers lightly touch her temple. Dress her in a plain neutral baseball cap without lettering, a structured dark-olive jacket over a light-beige T-shirt, tailored charcoal chinos, minimal off-white leather sneakers, a thin chain necklace, one simple ring and thin-metal round eyeglasses. The necklace is a styling item specified by the source text, not an object extracted from the face reference. Show smooth stone pavement with subtle reflections, a metal railing along the river, softly focused city buildings and an overcast sky with diffused daylight. Use soft natural side light, gentle facial shadows, a moody yet clean editorial atmosphere, muted earthy grading, soft contrast, clear subject detail and natural skin texture. Use a 50mm-prime, f/2.8, eye-level slightly side-on photographic look as visual targets, not a claim of actual camera hardware or settings. Keep the full figure and shoes in frame while retaining readable environmental context. Avoid identity drift, text, watermarks, extra people, distorted anatomy, harsh contrast and an excessively blurred environment.

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