OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image
Original Subject with Memphis-Style Split Illustration
Original Subject with Memphis-Style Split Illustration — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.
Prompt
Chinese prompt
创作一张经过精心设计的竖版上下双拼插画，上下两区严格等高，不做拼贴。必须使用用户上传的原图作为唯一主体来源。上半部：识别一个明确核心主体，例如主要人物、猫、狗或其他常见宠物；重点呈现主体，不让建筑和环境成为主角。保留主体识别特征：若为人物则保留其脸型、五官、发型、服饰与气质；若为宠物则保留品种、毛色花纹、耳朵和体型。主体完整清晰，背景适度简化。若核心主体为人物，只呈现一位明确成年、20岁以上的人物，使用固定身份ref-10a-v1，不新增其他人物；若为动物，则准确保留原动物身份，不套用人类参考。下半部：将同一主体转译为可辨识的孟菲斯视觉语言，融入波点、锯齿、小型几何图案、鲜艳配色与80年代玩趣。采用清晰几何结构、平面形状、有序构图、明确层级与充足留白。将主体轮廓、姿态、服饰或毛发花纹及负空间与孟菲斯装饰几何和有节奏的图形点缀融合。下半部必须一眼看出是同一主体，而不是另一个对象。全图保持协调统一、自然手作感与鲜明孟菲斯设计特征；避免混入无关画风、过度平滑、无意义细节堆砌或随机装饰。除非主题明确需要，不添加文字、标志、水印、边框或UI；不要多图拼接。
English prompt
Create a carefully designed portrait-oriented split illustration with two strictly equal-height sections, not a collage. The uploaded original image is required and is the sole source for the main subject. Upper section: identify one clear focal subject, such as the principal person, cat, dog or another common pet; preserve the subject rather than making buildings or surroundings dominant. Retain the subject's identifying features, hairstyle/clothing/temperament if human, or breed/fur color and patterns/ears/body shape if a pet; keep the subject complete and the background simplified. If the focal subject is human, depict one clearly adult person age 20+ using fixed identity ref-10a-v1, and do not add other people. If it is an animal, preserve that exact animal and do not apply a human identity. Lower section: reconstruct the same subject in recognizable Memphis visual language: polka dots, zigzags, small geometric motifs, vibrant color and playful 1980s energy. Use clear geometric structure, flat shapes, orderly composition, controlled hierarchy and generous breathing space. Integrate the subject's silhouette, pose, clothing or fur patterns, and negative space with decorative Memphis geometry and rhythmic graphic accents. The lower section must immediately read as the same subject, not a different object. Keep one coherent style, a natural handmade feel and recognizable Memphis design; avoid unrelated styles, over-smoothing, meaningless detail piles and random decoration. Unless the theme specifically requires them, omit text, logos, watermarks, borders and UI; no multi-image stitching.
Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas
Related works
- Create a realistic, lightly comic snapshot in an ordinary university
- Natural Light Portrait of an Adult Woman at a Modern Japanese Tea Setting
Natural Light Portrait of an Adult Woman at a Modern Japanese Tea Setting — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.
- English Landscape Poster with Flowing Contours and Bold Colors
English Landscape Poster with Flowing Contours and Bold Colors — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.
- Close-up Japanese portrait of a figure in a white yukata before a paper door
Close-up Japanese portrait of a figure in a white yukata before a paper door — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.
- Portrait of a Bohemian Woman in a Morning Courtyard
Portrait of a Bohemian Woman in a Morning Courtyard — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.
- Japanese Night Fashion Portrait in a Warmly Lit Bedroom
Japanese Night Fashion Portrait in a Warmly Lit Bedroom — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.