OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image
夜晚豪车后座的直闪狗仔风西装抓拍
Paparazzi Flash Candid of a Suit in the Back Seat of a Luxury Car at Night
一张夜晚豪车后座的近景时尚抓拍：成年女性穿黑色细条纹西装与白衬衫、黑领带，在直闪下自然轻笑；黑色车厢、手持构图、清晰真实的肌肤纹理和闪光阴影营造编辑式狗仔照片观感。面部身份使用指定参考图。
提示词 / Prompt
中文提示词
以用户提供的人物参考图作为画面中唯一可见成年女性的面部身份依据。严格保持参考图中的脸型、眼睛、鼻子、嘴唇、肤色与面部比例，不对身份特征做改动；仅按以下方向安排妆发、服装、姿态、构图和环境。 创作一张夜晚豪华轿车后座的真实抓拍式时尚人像。主体是一位明确28岁的成年女性，穿定制黑色细条纹西装、肩线利落的廓形外套、挺括白衬衫、修身黑色领带与黑色西裤，配小银色耳圈、精致手链和腕表。妆容精致克制，嘴唇带光泽，眼妆清晰，柔和修容；头发盘成略显随意但优雅的发髻，几缕发丝自然落在脸旁。 采用近景肖像与略宽的自然透视，呈现车内夜间抓拍的手持感。她在自然发笑，眼睛半闭，头部微微向后倾，情绪轻松自发、没有摆拍感；闪光瞬间凝住动作，几缕发丝轻微飞起。黑色真皮后座与车窗外的昏暗环境作为背景，保持简洁。使用强烈直射正面闪光，形成清晰高光、硬朗柔化边缘的阴影、高对比度和真实闪光衰减；皮肤高光可略亮，但仍保留可见毛孔、细纹和自然瑕疵，不磨皮、不做塑料质感。浅景深、轻微胶片颗粒与略不完美的手持取景强化编辑式狗仔时尚照片氛围。无其他人物、无文字、无水印。
English prompt
Use the user-provided identity reference for the facial identity of the single visible adult woman. Strictly preserve the reference face shape, eyes, nose, lips, skin tone and facial proportions; do not alter identity features. Apply the styling, clothing, pose, framing and setting below without changing that identity. Create a candid fashion portrait in the back seat of a luxury car at night. The subject is a clearly 28-year-old adult woman wearing a tailored black pinstripe suit, a structured blazer, crisp white shirt, slim black tie and black tailored trousers. Add small silver hoop earrings, a delicate bracelet and a watch. Keep makeup refined and restrained, with glossy lips, defined eyes and soft contour. Style her hair in a slightly messy but elegant updo with a few loose strands framing her face. Use a close-up portrait with a natural slightly-wide perspective and a handheld candid feel. She is laughing softly in a spontaneous, unposed moment, eyes half-closed and head tilted slightly back; freeze the instant with a few hair strands caught in motion. Use black leather car seats and a dark exterior beyond the windows as a simple night backdrop. Apply strong direct frontal flash for crisp highlights, firm shadows with natural edge softness, high contrast and realistic flash falloff. Skin highlights may be bright but retain visible pores, fine lines and natural imperfections—no skin smoothing or plastic texture. Shallow depth of field, subtle film grain and slightly imperfect handheld framing complete the editorial paparazzi-fashion look. No other people, text or watermark.
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
- 奢华美妆香水广告大片
以已配置的人像作为女性身份参考，创作奢华香水美妆广告；保留“Timeless Beauty”文字、Dior风格灵感香水瓶和柔和渐变光线。 Create a luxury perfume beauty advertisement using the configured portrait as the woman’s identity reference; retain the “Timeless Beauty” text, Dior-inspired bottle design and soft gradient lighting.
- 温暖日系可爱角色设定表
以已配置的人像作为女性角色的身份参考，制作温暖柔和的半写实日系卡哇依角色设定图，包含主肖像、表情、动作姿势、物件图标、调色板、虚构姓名与个性档案。 Create a warm, soft, semi-realistic Japanese/kawaii character sheet using the configured portrait as the woman character’s identity reference, with a main portrait, expressions, action poses, object icons, palette and fictiona
- 都市咖啡馆黑色西装时尚封面
以本次已配置的人像作为女性主角的身份参考，创作都市咖啡馆中的高级时尚杂志封面；保留4:5构图、黑色西装、黄金时刻光线及全部指定封面文字与条形码。 Create a premium fashion-magazine cover in a warm urban café, using the configured portrait as this preview’s woman identity reference and preserving the 4:5 layout, black blazer, golden-hour lighting, all specified cover text
- 《千里江山图》意象青绿丝巾
设计一条以《千里江山图》为意象、由王希孟青绿山水风格环绕的丝巾；表现丝绸质感与柔和光线，不宣称精确复刻原画。 Create a silk scarf design evocative of A Thousand Li of Rivers and Mountains, surrounded by Wang Ximeng’s blue-green landscape aesthetic, with gentle silk texture and soft light; do not claim an exact painting reproduction.
- 3D卡通少女表情贴纸组
以已配置的人像作为统一角色身份参考，创作同一位年轻女性的3D卡通表情贴纸；完整保留4×4共16种标签、表情与动作。 Create a glossy 3D cartoon emotion sticker pack of the same young woman, using the configured portrait as the identity anchor and preserving all 16 specified labels, expressions and poses in a 4x4 grid.
- 十二宫格人物发型设计大赏
以已配置的人像作为角色身份参考，设计十二种不同发型，并分别搭配服装与姿势，以十二格展示；为每格提供造型适配度评分。 Use the configured portrait as the character identity reference to create twelve hairstyle variations with coordinated clothing and poses in a twelve-cell grid, each with an individual styling-suitability score.