OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image
曼谷耀华力路夜景回眸人像
Portrait Looking Back on Yaowarat Road at Night, Bangkok
9:16曼谷耀华力路夜间街拍：成年女性持泰式奶茶回眸微笑，配参考服装、夜市霓虹和嘟嘟车；面部身份使用ref-10a-v1。
提示词 / Prompt
中文提示词
以用户上传的参考图作为主要可见人物的面部身份依据。在不与身份参考冲突的情况下，保留来源的场景、造型、服装、姿势、构图、光线和视觉风格。 创作一张9:16竖幅、色彩鲜明且超写实的曼谷唐人街耀华力路夜景街拍。主要人物是一位明确成年的女性；仅使用ref-10a-v1作为面部身份依据。面部保持自然，不经风格化改造，并摘掉眼镜；保留真实面部结构与皮肤质感。采用半身至四分之三身构图，同时展示关键服装和配饰。她背对或略侧背朝向镜头，展示发型与牛仔裤，再越过左肩转头直视镜头，露出明朗、友善、俏皮的笑容；腰肩自然扭转。左手举起一杯带吸管的透明塑料杯泰式奶茶，右手放松垂在身侧。 头发为浓密、笔直的深棕色长发，几缕碎发随风飘动，少量发丝自然掠过脸侧。妆容清新水润，深棕色眼睛、根根分明的卷翘睫毛、柔和粉色腮红和粉色渐变唇彩；避免塑料皮肤或过度油亮。身材比例自然，时尚呈现得体、不露骨。若用户在使用时提供服装照片，仅将其用于服装细节，不用于面部身份；若未提供，则按文字描述搭配：黑色单肩包，挂有Pop Mart Cry Baby Sunset Concert毛绒钥匙扣；黑色甲油胶指甲；左腕佩戴银色表壳、黑色表盘的GUESS手表；颈部佩戴极小的黑色四叶草吊坠项链。 场景设在耀华力路繁华夜市，深色夜空衬托红、绿、黄霓虹招牌。显眼背景包含沿街排列、文字层叠的高大竖向中文霓虹灯。在画面右上方放置两处准确且可读的招牌文字：“Hotel Royal Bangkok @ Chinatown”和“Hua Seng Heng”。街边停放数辆带装饰灯的蓝绿色嘟嘟车，远处有一辆略微虚化的黄绿色出租车。加入柔和虚化、正在行走的人群，不要抢走主体注意力；人物站立位置旁设置金属行人隔离栏。使用直接正面光或轻微闪光灯效果，让人物在夜景中明亮突出，同时保留真实阴影与色彩。面部、头发、妆容和衣料细节清晰，整体保持自然高端街拍质感。避免改变人物身份、幼态或未成年形象、眼镜、夸张解剖、多余或粘连手指、杯子或吸管变形、手臂融合、服装与身体穿模、僵硬姿势、地标文字错乱、额外标志、水印、强HDR、过度锐化、卡通、CGI、清晰度不足或影棚感。
English prompt
Use the uploaded reference image as the facial identity reference for the main visible person. Keep the source scene, styling, clothing, pose, composition, lighting and visual style unless they conflict with the identity reference. Create a vibrant, ultra-realistic 9:16 vertical night street photograph on Yaowarat Road in Bangkok Chinatown. The main subject is one clearly adult woman; use ref-10a-v1 as the only facial identity reference. Keep her face natural and unchanged by stylization, remove glasses, and retain realistic facial structure and skin texture. Frame her as a half-body/three-quarter portrait while showing the key outfit and accessories. She stands back or slightly side-back to camera to show her hairstyle and jeans, then turns over her left shoulder to look directly into the lens with a bright, friendly, playful smile. Twist her waist and shoulders naturally. Her left hand raises a clear plastic cup of Thai tea with a straw; her right hand rests relaxed at her side. Give her very long, thick, straight dark-brown hair, with a few loose strands moving in the breeze and a small natural strand falling across her face. Use a fresh dewy makeup look with deep-brown eyes, separated curled lashes, soft pink cheeks and pink gradient lip gloss, while avoiding plastic skin or excessive shine. Keep the figure naturally proportioned and the fashion presentation non-explicit. If the user supplies an outfit photo at use time, use it only for clothing details, never for facial identity; if absent, follow this described outfit: a black shoulder bag with a Pop Mart Cry Baby Sunset Concert plush keychain, black gel nails, a GUESS watch with silver case and black dial on the left wrist, and a tiny black clover pendant necklace. Set the scene in lively Yaowarat night traffic, with a dark sky and vivid red, green and yellow shop neon. Tall vertical Chinese neon signs with stacked characters form the prominent background. In the upper-right background, show the exact readable signs "Hotel Royal Bangkok @ Chinatown" and "Hua Seng Heng". Park several blue/green tuk-tuks with decorative lights along the street and place a yellow-green taxi far behind, slightly blurred. Add a softly blurred moving pedestrian crowd that does not compete with the subject, plus a metal pedestrian barrier beside where she stands. Use direct frontal light or a subtle camera-flash effect to make her bright against the night while preserving realistic shadows and color. Keep face, hair, makeup and clothing fabric detailed and the street-photo aesthetic candid and high-end. Avoid changing identity, youthful/underage appearance, glasses, exaggerated anatomy, fused or extra fingers, distorted cup or straw, merged arms, clothing-body intersections, stiff pose, garbled landmark text, extra logos, watermark, strong HDR, oversharpening, cartoon, CGI, low clarity or studio appearance.
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
- GPT-Image-2.5 像素解构主义
以上传的实拍照片作为唯一素材源，制作完整竖向构图作品，禁止单独生成抽象画、禁止直接给原图加滤镜 画面上下分区，上方为原图摄影区域，占画面50%，完整保留原始照片内容:下方将上传照片转换为“像素消隐景观“风格 插画。保留原照片中最具识别性的主体、基本构图、透视关系和主要色彩氛围,将真实场景提炼为由少量方块组成成的极 简像素化景观。不要直接复制照片细节,而是将建筑、街道、树木、人物、动物、交通工具或其他主体概括为简洁的几何 块面,用规则小方块重新组织画面,呈现一种照片被数字化、提纯后逐渐显现又逐渐消散的视觉效果。严格控制像素方块总 量,整体不超过500个方块。其中主体核心区域使用较密集的方块构成,
- 河边
清澈河水中的写实水下第一人称视角：左下近景有一条橘红鱼，镜头贴近河床仰望岸边石台上并坐的三位成年古风女子。三人分别穿月白、藕荷、青绿色交领襦裙，挽起裙摆避免浸湿衣料，深V领口，丰满身材，双腿与脚浸入浅水；水草、卵石、气泡、折射和午后光柱营造自然戏水氛围。
- H3爆裂冲击帧之【复仇之刃】
H3最强特效生成 粉丝福利邀请码：rh-v1631 10 15 33 三合一 SD2.5之下皆蝼蚁 全网最干净方便的空间折叠工作流，去除繁复连线、虚空桥接+空间折叠； B站主页：https://space.bilibili.com/111985620/lists?sid=7895659&spm_id_from=333.788.0.0 youtube主页：
- 极简双色品牌吉祥物头像与字标
9:16 极简扁平品牌视觉：纯色背景中央放置小型白色反白吉祥物头像，内部以底色镂空形成五官、刘海和发丝分界；底部居中为微型白色品牌字标，右侧边缘另有极小 voxcat 字样和一颗青金石蓝圆点。模板保留主体与品牌名变量，示例默认使用抽象吉祥物头像和来源中出现的 voxcat 字样。
- 镜压成字实验展览海报
9:16 实验性展览海报模板：四个不等大的中文标题字从四角向中心压入，中央镜面金属通过反射切分并重组部分笔画，让材质真实参与字体与版面变化。用深蓝黑背景、暖象牙白文字、镜面银色和少量朱红强调色，保持清楚的三层阅读顺序、精细编辑排版与可读留白。
- 服装店内活力双V手势街拍人像
35mm 街拍风格全身肖像：一位 20 岁出头的成年东亚女性在明亮现代服装店内戴黑框眼镜微笑，双手在头侧比出 V 字，右腿弯曲向后踢起、左腿支撑身体。她穿黑色白条纹五角星针织毛衣、深灰瑜伽裤、白袜和米白厚底运动鞋，斜挎小型棕色包；橡木地板、折叠毛衣货架与柔和商业顶灯构成真实生活方式场景。