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

黑暗展厅复古日式壁画肖像

Dark Gallery Retro Japanese Mural Portrait

以上传参考图保持成年主体的面部、发型、姿态和轮廓一致，在黑暗展厅中以右上至左下的聚光光束、复古日式壁画和电影负空间构成孤独艺术人像。

黑暗展厅复古日式壁画肖像 — AI image creation
黑暗展厅复古日式壁画肖像

提示词 / Prompt

中文提示词

使用上传参考图作为唯一身份参考，保持明确成年主体的面部、发型、刘海、面部比例、肤色、身体姿势和整体轮廓一致，不换脸、不漂移身份。创作 PAP 风格的黑暗展览室电影感写实肖像：主体独自站立，穿复古白色衬衫、粉色外衣和黑色复古材质裙，极简妆容；身体面向相机但呈约70度侧面角度，姿势放松，一个肩膀略低，双手整理头发，神情冷静神秘，具有智慧优雅气质。主体位于画面中下方，被大片黑暗包围；一道来自右上方、斜向左下方的强聚焦聚光灯切过场景，照亮身体约一半，其余自然淡入深影，脸部和眼镜边缘留柔和暖色高光。身后是带复古卡通插图与传统艺术细节的彩色日本复古壁画，前景阴影像撕裂的黑暗一样自然框住画面，形成电影感负空间与忧郁博物馆氛围。平视机位，中近景环境人像，与主体保持轻微距离；低光摄影，浅景深，真实阴影衰减，细微胶片颗粒、模拟质感、Kodak Portra式色彩，静谧孤独的艺术电影剧照与低调高级时尚编辑感。保持自然皮肤纹理和细微瑕疵。负向提示：动漫、微笑、过曝、多余肢体、错误解剖、鲜艳现代服装、赛博朋克霓虹、模糊脸、低质量、扭曲手部、不真实皮肤、HDR过载、夸张妆容、构图杂乱、重复人物、文字、徽标、水印。

English prompt

Use the uploaded reference as the sole identity reference, preserving the clearly adult subject's face, hairstyle, bangs, facial proportions, skin tone, body posture, and overall silhouette; do not face-swap or drift identity. Create a PAP-style cinematic realistic portrait in a dark exhibition room. The subject stands alone in a vintage white shirt, pink outer layer, and black vintage-textured skirt with minimal makeup. Her body faces the camera while captured from about a 70-degree side angle; posture relaxed, one shoulder slightly lowered, both hands adjusting her hair, expression calm and mysterious with an intelligent elegant aura. Place her in the lower center surrounded by darkness. A hard focused spotlight cuts diagonally from upper right to lower left, illuminating roughly half her body while the rest falls naturally into deep shadow; leave a soft warm highlight on her face and the rim of her glasses. Behind her is a colorful vintage Japanese mural wall with retro cartoon illustrations and traditional art details. Let foreground shadows frame the image like torn darkness, creating cinematic negative space, a moody museum atmosphere, solitary artistic feeling, and contemplative indie-film still energy. Use an eye-level camera, medium-close environmental portrait, slight distance for storytelling, low-light photography, shallow depth of field, realistic shadow falloff, subtle film grain, analog texture, and Kodak Portra-like color. Keep natural skin texture and subtle imperfections with understated luxury editorial styling. Negative prompts: anime, smiling, overexposure, extra limbs, bad anatomy, bright modern outfit, cyberpunk neon, blurry face, low quality, distorted hands, unrealistic skin, HDR overload, exaggerated makeup, cluttered composition, duplicate person, text, logo, 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