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

原片与纸艺

Original Photographs and Paper Art

为每张输入照片单独制作3:4竖版社论海报：上半部保留原片，下半部以分层剪纸重构其主要轮廓与空间关系，并以克制的图文层级完成统一设计。

原片与纸艺 — AI image creation
原片与纸艺

提示词 / Prompt

中文提示词

为每张上传的照片分别制作一张精致的社论风格海报，绝不把多张照片拼成一张。画布严格为竖幅3:4，水平方向分成上下两个完全相等的部分，上下各占50%。上半部分——保留所提供的原始照片及其主体不变，维持真实纹理、自然光线和原有色彩氛围；只进行细微的艺术杂志或展览摄影级色彩校正。如需适配构图，只可轻轻延伸周围天空、地面或环境背景；绝不拉伸、扭曲或改动主体。下半部分——将照片中最具辨识度的主要主体、轮廓、姿势和空间关系重新诠释为极简分层纸艺。以一个视觉主体为锚点，通过清晰的折叠、裁切、分层和交叠重构关键形状与比例，而非逐一照搬所有细节。只把少数辅助环境元素转化为前景、中景和背景纸层；利用比例对比、负空间、精确对齐和清楚的前后层次，让原图主体仍能立即辨认，同时保持克制的社论构图。从原照片提取鲜明色彩并简化为有限、和谐的纸艺配色：暖象牙色或浅色卡纸为底，少量较强强调色、较深结构色、柔和过渡色和极少高光。材质表现为高品质哑光卡纸，带有细微可见的纸纤维、整洁切边，以及自然漫射光下柔和的接触阴影。避免塑料般的亮面3D效果、廉价手工纸风格、霓虹色，以及过饱和或复杂渐变。根据图像主体的身份、地点、动作、材质、情绪或象征意义拟定一个简短主标题，并添加2–4组有意义的小字，例如物体名称、地点、序列、方向、状态、材质、分类或极短诗意短句。不得虚构年份或日期。让文字与纸艺自然融合，可沿纸层边缘、主体轮廓、几何轴线或负空间排布；可采用横排或竖排、旋转、宽松字距、边缘对齐、角落放置、横截面布局、嵌入式排版或局部交叠。主标题可带细微剪纸或折纸质感，辅助文字保持纤细、现代、克制且精确。所有文案都须有意图并清晰可读，不使用 Lorem Ipsum 或随机填充文字。成品应优雅、富有触感、平衡、经过细致艺术指导，并具有鲜明的社论风格。

English prompt

For each uploaded photo, create one separate refined editorial poster; never combine multiple photos into a collage. Use a strict vertical 3:4 canvas split horizontally into two exactly equal sections, top and bottom each occupying 50%. TOP HALF — preserve the supplied original photograph and its main subject unchanged, retaining realistic texture, natural lighting and original color atmosphere. Apply only subtle art-magazine or exhibition-photography color correction. If needed, gently extend only the surrounding sky, ground or environmental background to fit; never stretch, distort or alter the main subject. BOTTOM HALF — reinterpret the photo's most recognizable primary subject, silhouette, pose and spatial relationships as minimalist layered paper art. Use one visual anchor and reconstruct its key shape and proportions through clean folds, cuts, layers and overlaps rather than copying every detail. Convert only a few supporting environmental elements into foreground, midground and background paper layers. Use scale contrast, negative space, precise alignment and clear depth so the source remains immediately recognizable within a restrained editorial composition. Extract the source photo's distinctive colors and simplify them into a limited harmonious paper palette: warm ivory or light cardstock base, a small amount of stronger accents, darker structural tones, soft transition colors and very limited highlights. Show high-quality matte cardstock with subtle visible fibers, clean cut edges and soft contact shadows under natural diffused light. Avoid glossy plastic-like 3D, cheap craft-paper styling, neon, and oversaturated or complex gradients. Add an image-derived short main title that expresses the subject's identity, place, action, material, emotion or symbolism, plus 2–4 small groups of meaningful supporting text such as object names, location, sequence, direction, status, material, classification or a very short poetic phrase. Do not invent a year or date. Integrate the typography with the paper forms along layer edges, contours, geometric axes or negative space; horizontal or vertical setting, rotation, generous tracking, edge alignment, corner placement, cross-section layout, embedded type or partial overlap are acceptable. The title may have a subtle cut/fold-paper quality; supporting type stays thin, modern, restrained and precise. Keep all copy intentional and legible, with no Lorem Ipsum or random filler. The result should feel elegant, tactile, balanced, carefully art-directed and distinctly editorial.

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