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

都市蒙太奇拼贴风格的夸张人物肖像

Exaggerated Portrait in Urban Montage Collage Style

保留参考人物身份特征的都市照片拼贴漫画肖像，结合街景碎片、纸张层次和手绘笔触。

都市蒙太奇拼贴风格的夸张人物肖像 — AI image creation
都市蒙太奇拼贴风格的夸张人物肖像

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。 创作一幅4:5竖幅都市照片拼贴漫画肖像，主体为a clearly adult person from the uploaded identity reference。人物面部保持清晰可辨，同时通过头部轮廓、眉眼和动作做有表现力的编辑式夸张，但不扭曲身份。画面由匿名城市环境的分层碎片构成，例如人行横道条纹、店铺橱窗、砖墙和类似交通图的几何元素，并结合撕纸边缘、胶带照片角、墨线轮廓与随性的马克笔涂画。这些都是图形拼贴元素，不代表可辨认的真实照片。 将成年人物的插画面部与肩部作为主要焦点层，穿钴蓝色外套和简洁中性色上衣。采用鲜明的钴蓝、暖橙、奶油色和炭灰配色，前景与背景层次明确。拼贴碎片保持均衡、易读，不出现可辨招牌、标志、品牌标记或额外文字。成品应充满活力、富有手作感，并清晰呈现都市气息，将漫画绘制与混合媒材拼贴结合起来。

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 vertical 4:5 urban photo-collage caricature portrait of a clearly adult person from the uploaded identity reference. Keep the face clearly recognizable while using expressive editorial exaggeration in the head silhouette, brows and gesture, without distorting identity. Build the image from layered fragments of an anonymous city environment—crosswalk stripes, storefront windows, brick and transit-map-like geometry—combined with torn-paper edges, taped photo corners, ink contours and loose marker scribbles. These are graphic collage elements, not identifiable real photographs. Place the adult subject's illustrated face and shoulders as the dominant focal layer, wearing a cobalt jacket over a simple neutral shirt. Use a vivid palette of cobalt, warm orange, cream and charcoal, with clear separation between foreground and background. Keep collage fragments balanced and readable, with no legible signage, logos, brand marks or extra text. The result should feel energetic, handcrafted and distinctly urban, combining caricature drawing with mixed-media collage.

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