OpenZhenZhen AI Prompt Gallery · Moss Image · image

Goldfish Glass Teapot · Dreamy Soft Focus

Goldfish Glass Teapot · Dreamy Soft Focus — explore this Moss Image creation on OpenZhenZhen, with its original prompt and a reusable generation example.

Goldfish Glass Teapot · Dreamy Soft Focus — AI image creation
Goldfish Glass Teapot · Dreamy Soft Focus

Prompt

Chinese prompt

参考图只用于核心主体轮廓与大体空间布局;请以下列新视觉语言完全重新诠释媒介、色彩、材质与照明,创作全新完整作品,避免近似复制。创作一幅出色的「梦幻柔焦」作品:透明玻璃茶壶形成安静水下景观;一尾金鱼游过细小沉水茶叶。视觉语言:珠光高光、细腻柔化、克制发亮光晕、真实触感与安静情绪纵深。构图:宽广构图配合微小尺度参照,赋予主体纪念碑般体量。画面应像经过精心艺术指导:确立单一引人入胜的焦点,组织有意义的微细节,建立尺度、材质与环境之间的协调关系。保持所选媒介的统一,不混入不相容的渲染风格。避免普通素材图式处理,仔细完成轮廓、边缘与空间纵深。全心刻画所列主体与环境,画面不添加人物、人影或人形。完整独立单幅,不要拼接面板、水印、品牌标志或可读文字。

English prompt

Use the reference only for the core subject silhouette and broad spatial arrangement; completely reinterpret its medium, colors, materials and lighting into the new visual language below. Make a fresh finished artwork, not a near-duplicate. Create an exceptional dreamlike editorial photography artwork depicting a transparent glass teapot shaped as a tranquil aquatic landscape; one goldfish swims between tiny submerged tea leaves. Visual language: pearlescent highlights, fine diffusion, subtle luminous bloom, tactile realistic materials and emotionally quiet atmospheric depth. Composition: a wide composition using a small scale cue to make the central structure feel monumental. Make the concept feel deliberately art-directed: establish a single compelling focal point, meaningful fine details and a coherent relationship between scale, materials and setting. Preserve the chosen medium rather than mixing incompatible rendering styles. Avoid generic stock imagery. Carefully resolve silhouettes, edges and spatial depth. Focus entirely on the listed subjects and setting; this scene contains no human figures or human silhouettes. One finished standalone image, no collage panels, no watermarks, no logos and no readable lettering.

Topics and styles

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas