Create a vertical "upper-lower split" style poster, overall like a disassembled picture book page: the upper part is a real photo, and the lower part is a fairytale print-style image translation of the same subject. The composition is simple, clear, narrative, with the temperament of a picture book and an art publication. Upper Part: Display the original photo uploaded by the user. Retain the main characters, pets, animals, or the most important interaction between characters and animals in the photo. The subject must be clearly identifiable, trying to preserve its identity, movement, posture, expression, clothing features, fur color, outer contour, and emotional state. The upper part maintains a sense of real photography and does not undergo painterly processing. Environments and buildings exist only as auxiliary spatial clues and do not occupy the main position. Lower Part: Extract the most recognizable subjects, outlines, postures, and narrative relationships from the upper photo, and reconstruct them into fairytale print-style hand-drawn images. Do not mechanically trace details, but instead compress the subject into one or two clear large blocks, using blunt silhouettes, slightly exaggerated proportions, and a few key edges to retain the original object's identity, action, function, and emotion. The image refines a large area of low-brightness or high-contrast structural field from the original content, serving as a container, path, frame, or density layer to wrap the subject; the background grows out of the original space and meaning like a shell where the story takes place, forming a structural narrative of blocking, cropping, swallowing, and partial exposure. The overall style adopts flat narrative perspective and picture book page-style hierarchy, utilizing overhead compression, lateral movement, front framing, path guidance, scale jumping, or edge cropping to organize reading. Color: Extracted from the uploaded original image, maintaining a clear three-layer relationship: large areas of dark structural colors are responsible for weight and boundaries, appropriate amounts of light colors or negative space are responsible for breathing, and small areas of high-saturation emotional colors are responsible for the image reading path and story signals. The most important information is concentrated only in a few bright color blocks on the face, posture, path, utensil function points, environment markers, or narrative focus, and is always suppressed by dark structures or light negative shapes, making the image vivid but not balanced. The hue can be adjusted according to the theme's mood, but the area ratio and brightness order remain stable. Lines: Appear only at functions, paths, boundaries, textures, and identification points, maintaining sparse, hand-drawn, slightly inaccurate thin lines. All color blocks maintain flat processing, while retaining dry brush, grain, white space (feibai), paper texture, spray dots, and slight overprint misalignment, presenting a handcrafted texture completed by screen printing and pastels, avoiding realistic lighting and shadow, complex volumes, smooth vector edges, 3D rendering, and plastic textures. Overall: The composition is simple, low-detail, high-recognition, with a dark field wrapping bright signals, combining cute narrative with a heavy print feel. If the image contains text, the text can be treated as small logos, object panels, or edge annotations, with a streamlined quantity that subordinates to the graphic structure; the font is round and blunt, with slightly uneven thickness and low density, carrying a natural handwritten feel, making it look like a part of the image rather than an explanation pasted on later. Overall Presentation: A visual relationship of "upper part real original image + lower part fairytale picture book, handcrafted print, editorial illustration, and art publication temperament combination," forming a very clear "before and after translation of the same subject" effect, with unified, vivid, and transferable visuals. 创作一张竖版“上下双拼”风格海报,整体像一本被拆开的绘本页:上方是真实照片,下方是同一主体的童话版画式图像转译。构图简洁、清楚、带叙事性,具有绘本和艺术出版物气质。 上半部分: 展示用户上传的原始照片。保留照片中的主要人物、宠物、动物,或人物与动物之间最重要的互动关系。主体必须清晰可辨,尽量保留其身份、动作、姿态、表情、服装特征、毛色、外轮廓与情绪状态。上半部分保留真实摄影感,不进行绘画化处理。环境和建筑只作为辅助空间线索存在,不占据主体地位。 下半部分: 从上半部分照片中提取最具识别性的主体、轮廓、姿态与叙事关系,重构为童话版画式手绘图像。不要机械描摹细节,而是将主体压缩为一两个清楚的大体块,以钝拙剪影、略微夸张的比例和少量关键边缘,保留原始对象的身份、动作、功能与情绪。画面从原图内容中提炼出一个大面积低明度或高反差的结构场,作为包裹主体的容器、路径、框架或密度层;背景像故事发生的外壳一样从原图空间与意义中生长出来,形成遮挡、裁切、吞没与局部露出的结构叙事。整体采用平面叙事透视和绘本页面式层级,可使用俯视压缩、侧向移动、正面框架、路径引导、尺度跳跃或边缘裁切组织阅读。 配色从上传的原图提取,并保持明确的三层关系:大面积暗结构色负责重量与边界,适量浅色或留白负责呼吸,小面积高饱和情绪色负责读图路线与故事信号。最重要的信息仅以少量鲜明色块集中在面部、姿态、路径、器物功能点、环境标识或叙事焦点上,并始终被深色结构或浅色负形压住,让画面鲜艳但不平均。色相可随主题情绪调整,但面积比例与明度秩序保持稳定。 线条只在功能、路径、边界、纹理和识别处出现,保持稀疏、手绘、略不准的细线。所有色块保持平面化处理,同时保留干刷、颗粒、飞白、纸纹、喷点和轻微套印错位,呈现丝网印刷与粉蜡笔共同完成的手工质感,避免写实光影、复杂体积、光滑矢量边缘、三维渲染和塑料质感。 整体构图简洁、低细节、高识别,暗场包裹亮色信号,兼具可爱叙事与厚重版画感。若画面包含文字,可将文字处理为小标识、物体面板或边缘注释,数量精简,服从图形结构;字形圆钝,粗细略不均,密度低,带有自然手写感,使其像画面的一部分而不是后期贴上去的说明。 整体呈现“上半部分真实原图 + 下半部分童话绘本、手工版画、编辑插画与艺术出版物气质结合”的视觉关系,形成非常明确的“同一主体前后转译”效果,视觉统一、鲜明、可迁移。

{ "project_configuration": { "version": "2.0", "output_quality": "Ultra-HD 8K", "aspect_ratio": "3:4", "render_engine_hints": [ "パストレーシング", "サブサーフェス・スキャタリング", "物理ベースレンダリング(PBR)" ] }, "subject_profile": { "model": "{argument name=\"モデル名\" de
