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

空间透视建筑体量排版海报

Spatial Perspective Architectural Mass Layout Poster

空间透视建筑体量排版海报，使用上传人物参考图绑定主可见人物身份。

空间透视建筑体量排版海报 — AI image creation
空间透视建筑体量排版海报

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。创建一张9:16高冲击力透视排版海报，让超大号字体在物理上融入建筑，并仅在一个预定相机视角下变得完整。先构建建筑结构，再让排版严格遵循真实深度、表面、角度、缝隙、折叠和遮挡。使用切片、跨度、偏移或折叠平面，让碎片从预定视角对齐成一个清晰可读的词。主要词汇约占构图35-55%，使用粗体几何无衬线字体、一种纯色强调色和克制的混凝土、黑、白或中性石材。若包含人物，保持其微小并作为比例参考，让建筑自然遮挡或重叠人物。仅添加少量编辑信息。避免悬浮文字、随机扭曲、对齐不佳、装饰性挤压、霓虹科幻效果、过多文案、杂乱背景、水印和编号。

English prompt

Use the uploaded reference image as the facial identity reference for the main visible person. Create a high-impact 9:16 perspective typography poster where oversized type is physically integrated into architecture and becomes complete from one intentional camera viewpoint. Build the structure first, then make typography follow real depth, surfaces, angles, gaps, folds and occlusion. Use vertical slicing, spanning, offsets or folded planes so fragments align into one readable word from the intended view. Keep the main word at roughly 35-55% of the composition, use bold geometric sans-serif type, one solid accent color and restrained concrete, black, white or neutral stone materials. If a person is included, keep them small as scale reference and let architecture naturally overlap them. Add only minimal editorial information. Avoid floating text, random distortion, weak alignment, decorative extrusion, neon sci-fi effects, excess copy, clutter, watermarks and numbering.

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