OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image
山海经异兽低多边形二向色玻璃装置
Classic of Mountains and Seas Mythical Beasts Low-Poly Dichroic Glass Installation
每张选择一只《山海经》异兽，将其关键形态和结构数量转化为低多边形二向色玻璃空间装置；透射与反射色光落在简洁建筑空间，并投影一次准确中文名。
提示词 / Prompt
中文提示词
创作一张简洁空间艺术装置的真实摄影作品，装置灵感来自《山海经》中的一只异兽。本张独立选择一种轮廓鲜明、适合玻璃雕塑表达的异兽。保留其可辨认的体形，以及已有中文记载中的关键结构和准确数量；不要混合不同异兽，也不要凭空增加角、翼、尾或肢体。系列制作时，十张分别采用十种不同异兽和各异的主体结构。 将异兽塑造成连贯的低多边形玻璃雕塑，使用少量宽阔、层叠的二向色玻璃板。大面积表面保持简洁，仅在轮廓和主要结构转折处增加分面；以四边形及不规则多边形为主，只用少量三角面辅助转折。适用时保持躯干主体体块完整；肢体和附属结构以简练折面塑造，自然连接，并让关节、连接点和末端容易辨认。姿态符合异兽的解剖逻辑与重心，重叠轮廓之间留有清晰间隙。表现平整玻璃表面、清楚的薄板边缘、克制的接缝，以及顺应重心且可信的支撑或悬吊结构。 玻璃保持清透，使人能透过主体玻璃辨认中性的建筑背景；前后层叠关系清楚。颜色随面板朝向、透射和反射变化，并保留部分无色透明玻璃。清晰日光形成克制的彩色光影关系：透射光与反射光以宽阔、疏朗的色块落在地面或墙面，位置符合玻璃分面及光路，辅助反光轻淡。 设置一盏隐蔽的小型投影灯，配合文字字模和二向色滤色片，将所选异兽的准确中文名投射到墙面或地面的留白处，仅出现一次。使用端正易读的楷书，字幅适应空间，视觉分量低于雕塑。优先投向侧后方较暗的墙面，或透视变形较小的地面区域；文字顺应承接平面，清晰可辨、边缘略柔，并避开彩色主光块。不要添加其他文字、标签、Logo或水印。 采用简洁中性的建筑环境，空间尺度、背景明度和光线随雕塑调整。装置完整入镜，四周留出舒适边距，兼顾异兽辨识、玻璃构造、空间光影和单个中文名称的阅读。成片应像自然曝光下真实拍摄、可实际制作的艺术装置，玻璃层次与彩色光影清楚，雕塑始终是视觉主体。
English prompt
Create a realistic photograph of a concise spatial art installation inspired by one mythical creature from the Classic of Mountains and Seas (Shan Hai Jing). For this image, independently select exactly one creature with a distinctive silhouette suited to glass sculpture. Preserve its recognizable body form and the key structures and counts associated with its established Chinese description; do not combine creatures or invent extra horns, wings, tails or limbs. Across a series, vary the selected creature and the installation design so the ten images use ten different creatures and distinct main structures. Model the creature as a coherent low-poly glass sculpture using a small number of broad, layered dichroic-glass plates. Keep large surfaces calm; add facets mainly at the silhouette and major structural transitions. Favor quadrilaterals and irregular polygons, with only occasional triangular transition facets. Maintain a complete torso-like central volume where appropriate; shape limbs and appendages with simple folded planes that connect naturally and make joints, attachment points and endpoints easy to identify. The pose must follow plausible creature anatomy and balance, with clear gaps between overlapping silhouettes. Show flat glass surfaces, visible thin plate edges, restrained seams, and credible supports or suspension aligned with the center of gravity. Keep the glass transparent enough to recognize the neutral architectural background through it. Make front-to-back overlap legible and let color shift with panel orientation, transmission and reflection; preserve some clear, uncolored glass. Clear sunlight should create a restrained set of colored-light relationships: broad, sparse transmitted and reflected patches on the floor or walls whose positions agree with the glass facets and light path, plus only subtle fill reflections. Include one discreet small projector using a text stencil and dichroic filter to cast the selected creature's accurate Chinese name exactly once onto a blank wall or floor area. Use upright, legible regular script (Kaishu), scaled to fit the space and kept visually secondary to the sculpture. Choose a darker side/rear wall or a floor area with little perspective distortion; let the projected text follow the receiving plane, remain clear with slightly softened edges, and avoid the main colored-light patches. Do not add any other lettering, labels, logos or watermarks. Use a concise neutral architectural setting whose scale, brightness and light adapt to the sculpture. Keep the whole installation in frame with comfortable margins. Balance creature recognition, glass construction, spatial light and shadow, and readability of the single Chinese name. The result should look like a naturally exposed photograph of a physically buildable art installation, with crisp glass layers and colored light; the sculpture remains the visual focus.
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
相关作品
- 莱雷海滩黄金时刻电影感人像
泰国莱雷海滩黄金时段的超写实电影感齐腰旅拍肖像：以当前配置固定参考 ref-10a-v1 明确绑定模板中的第一人称“我”，人物穿宽松米白亚麻衬衫、经典太阳镜和简约腕表；背景为石灰岩悬崖及柔和虚化的碧绿安达曼海。完整保留自然发型、暖色肤调、海风、无修图质感及 85 mm 摄影要求。 | An ultra-realistic cinematic waist-up travel portrait at Railay Beach, Thailand: the currently configured fixed reference ref-10a-v1 is explicitly bound to t
- 吃我一脚
Photorealistic professional e-commerce photography, extreme low-angle frontal perspective directly facing the model's barefoot raised foot, the large sole and pointed toes fill the foreground with strong foreshortening, dramatic near-to-far perspective and visual pressure, elegant young woman with l
- 1
让材质更加干净整洁一些
- 漫步欧洲街头的三格时尚画报
垂直三格欧洲街拍时尚拼贴：以 ref-10a-v1 作为此可复用模板中女性主体的唯一面部身份参考；完整呈现街景、三种抓拍视角、服饰与配件、冰咖啡自拍及 Linkin Park 音乐播放器。面部特征按参考尽量保留，生成可能存在自然差异；64K 是高分辨率目标，不承诺实际像素尺寸。 | A vertical three-panel European street-style fashion collage using ref-10a-v1 as the sole facial identity reference for the female subject in this reusable te
- 俯拍夜幕下的便利店街头肖像
夜间街头时尚抓拍：从街对面较高楼层向下拍摄，一名年轻女子独自站在小型超市入口附近，略微偏离中心，呈现全身并保留开阔留白；穿白色 T 恤、水洗蓝色牛仔夹克、深色直筒牛仔裤和白色帆布运动鞋，单肩挂小手提袋。使用当前配置的固定面部参考 ref-10a-v1 作为本模板人物的面部参考。 | Candid night street-fashion scene viewed downward from an upper floor across the street: a young woman stands alone near a minimarket entrance, slightly off-c
- 最喜欢的角色进食动作
Add a food-specific eating action to the character in the configured reference image. Food is editable; defaults to a strawberry. The source mentions Nano Banana as its tool context. / 为配置参考图中的角色添加指定食物的进食动作；食物可编辑，默认一颗草莓。来源将 Nano Banana 作为工具语境提及。