OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image
朱墙银杏之间的秋日国风时尚组照
Autumn Chinese-Style Fashion Portraits Among Red Walls and Ginkgo Trees
一张4:5竖幅六格时尚编辑组照，呈现成年女性身着朱红长裙置身红墙与金色银杏之间，并在顶部展示准确品牌字样。
提示词 / Prompt
中文提示词
创作一张4:5竖幅写实时尚编辑组照，采用整洁的3列×2行六格网格。六格均为同一位明确成年的二十多岁东亚女性，并使用已配置的人物身份参考保持她的面部身份。她留红棕色蓬松大卷发，神情温柔而坚定，气质松弛克制；妆容可有电影感但不改变五官。她穿朱红色长裙，呈现天鹅绒或厚实棉麻的触感，搭配复古金色叠戴首饰，并披一条随秋风轻扬的轻薄半透明披肩。场景设在古建朱红墙、传统窗框与金黄银杏之间。A格：眼睛与上半脸的亲密特写，瞳中有含蓄的银杏光影。B格：45度侧脸近景，秋日逆光勾勒几缕卷发。C格：她在红墙窗沿放松侧坐，以手托腮。D格：纤细手指轻触脸颊、腕间叠戴金饰的细节特写。E格：红色古窗框景中的居中半身像，留有柔和空间。F格：捕捉银杏叶飘落的中景，人物清晰，前景落叶自然形成层次。六格保持人物身份、红裙、首饰与暖秋色调一致。采用柔和漫射的午后自然光、细腻金色轮廓光、适度胶片感颗粒，真实呈现毛孔、面料和斑驳红墙纹理；景深浅而人物面部清楚，背景柔焦。画面顶部仅放置一行小巧优雅的现代品牌字样“上城士 SHANG CHENG SHI”，不添加其他可读文字、说明、标志或水印。整体诗意、松弛、自信且具有触感，将中式框景构图与精致时尚编辑摄影结合。不要仿写具体摄影师、电影导演或品牌广告风格；不承诺4K/8K输出。
English prompt
Create a photorealistic 4:5 vertical fashion-editorial contact sheet with a clean 3-column by 2-row grid of six coordinated frames. Across every frame, preserve the same clearly adult East Asian woman in her late 20s and use the configured identity reference for her facial identity. Her reddish-brown large soft curls, gentle yet self-possessed expression, and elegant restrained presence remain consistent; cinematic makeup may be refined without changing her features. Style her in a vermilion-red long dress with tactile velvet or substantial cotton-linen texture, refined vintage gold layered jewelry, and a light translucent shawl flowing gently in the autumn breeze. Set the series among an old vermilion-red courtyard wall, a traditional window frame, and golden ginkgo leaves. Frame A: intimate eye-and-upper-face close-up with a subtle ginkgo reflection in the eyes. B: 45-degree side-profile close-up with autumn backlight along a few curls. C: relaxed seated pose on a red window sill, chin resting on one hand. D: detail of slender fingers touching her cheek and stacked gold bracelets. E: centered half-body portrait framed by the old red window, with soft negative space. F: mid-shot catching falling ginkgo leaves, keeping the woman crisp and foreground leaves naturally layered. Keep the six panels visually coherent, with consistent identity, dress, jewelry, and warm autumn palette. Use soft diffused afternoon light, gentle golden rim light, moderate film-like grain, realistic pores, fabric and weathered-wall texture, shallow depth of field, and a cinematic soft-focus background without losing facial clarity. Put exactly “上城士 SHANG CHENG SHI” in a small elegant modern lockup at the top; no other readable text, captions, logos, or watermark. The mood is poetic, tactile, relaxed, confident, and contemporary, blending Chinese framed-view composition with refined editorial photography. Do not imitate a named photographer, film director, or specific brand campaign; do not promise 4K/8K output.
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
相关作品
- 九十年代胶片感黑白街头男子肖像
Bilingual curation of the original black-and-white 1990s noir street-fashion portrait. The configured fixed reference is available for identity guidance; its adult-woman appearance is disclosed while the source subject remains male.
- 黄色阳台上的红衣男子抓拍肖像
Bilingual curation of the original 9:16 candid balcony portrait. The configured fixed reference supplies identity guidance; its adult-woman appearance is disclosed as a cross-gender tension while the depicted subject remains male.
- 暗黑黑手党奢华男模写真
OpenNana 原始双语时尚影棚人像提示词整理。保留 25–30 岁男性模特、橄榄色皮肤、精瘦运动员体格、乌黑后梳发、胡茬与胡须、指定三件套西装配饰、牙签姿势与纹身、酒红渐变影棚及电影级布光。面部身份绑定固定参考 ref-10a-v1；该参考呈现女性面孔，与来源男性呈现存在跨性别身份迁移张力，已在说明中保留。来源提出100%相似和8K；这里只作为尽力遵循与高细节风格意图，不保证实际结果或分辨率。
- 黄金时刻公路上的硬汉骑士
OpenNana 原始双语摩托车人像提示词整理。保留男性骑士、现代摩托车、单脚着地的自信坐姿、黑皮夹克、白圆领T恤、黑色现代框太阳镜、绿植环绕的开阔公路、远山与柏油路、黄金时段暖光和轮廓光、低角度略正面85mm浅景深构图。面部身份绑定固定参考 ref-10a-v1；固定参考呈现女性面孔，来源要求男性主体，已保留此呈现张力而不改性别。来源未指定画幅比例；不补造比例。来源提出8K，实际输出尺寸不作保证。
- 超写实白衬衫证件照
OpenNana 原始双语证件照提示词整理。保留护照尺寸/官方证件照风格、严格参照主体面部身份、白色领衬衫与配套领带、中性闭唇表情、正视镜头、双耳可见及包含肩部。两种来源语言均有残缺的类 JSON 片段；本版只规范化两种语言都清楚表达的内容，并保留“品牌现代风格”这一泛化含义，不补造品牌、标识或细节。提示表达来源的证件照意图，不保证实际政府证件适用性。
- 极简影棚手持艺术海报肖像
OpenNana 原始双语提示词整理。保留上传参考人物作为主体、手持同脸艺术再创作竖版海报、极简中性色影棚、柔和主光与轮廓光、从容自信姿态、浅景深与方形构图。海报艺术处理保留来源列出的数字插画、素描、绘画三种选择，默认数字插画。面部身份绑定固定参考 ref-10a-v1；高分辨率措辞不承诺实际工具输出尺寸。