neroidea
由 ChatGPT 根据所示提示词生成,点击图片可全屏查看。

真实照片与像素色彩索引对比海报

ChatGPT海报
提示词来自 @Hamburgerai

Use a single real photo as the sole source of content to create a vertical top-and-bottom comparison editorial poster. The upper half completely preserves the real photographic texture, subject identity, quantity, pose, spatial relationship, lighting, and composition of the input photo, without turning the photo into an illustration; the lower half is a stylized re-rendering area of the same photo: on warm white paper, the image is refined into four equal-sized square pixel color block indices, respectively corresponding to the most recognizable subject outline, architectural or object structure, relationship between main colors and light/shadow, and one environmental detail from the photo. Each grid uses hard-edged pixel blocks with a limited number of colors and a low-resolution grid feel, preserving traceable forms instead of random mosaics. The four grids are arranged in a 2×2 layout with ample paper space left around them; use thin black crosshair crop/registration marks to frame the index area, adding a very small amount of small bilingual titles and fields, such as "COLOR INDEX / SCENE STUDY," with restrained and clear text like print proof annotations that do not overshadow the main content. The lower half must look like a designed color index card, rather than a filter, ordinary pixelation, or repeated photo; do not add irrelevant objects or change the theme of the photo. The upper and lower areas each occupy about half, with a clear horizontal boundary; side-by-side layout is prohibited, multi-image collages are prohibited, "original/AI image" corner marks are prohibited, and watermarks and brand logos are prohibited. All colors are extracted and compressed from the input photo, and the content of the four grids changes according to the subject of the photo, applicable to any photo such as people, animals, architecture, street scenes, nature, interiors, food, and objects.

提示词解读

此海报将真实照片与像素风格对比呈现:上半页保留原照片的全部纹理与构图,下半则分成四个正方形像素块,分别对应主体轮廓、结构、光影关系与环境细节,形成低分辨率而有条理的色彩索引。 视觉冲击力来源于“保持上半页真实摄影纹理”和“下半页用硬边像素块再现”的两条核心指令。整体采用水平分隔、2×2 网格和细黑十字框标记,配以少量双语标题,强调内容转化而非装饰。 使用时只需上传任何照片,系统会自动生成对应的两个视觉面板。若想要黑白版本,只需让像素网格使用灰度调色方案;若需更卡通化,可在索引卡上加入微妙阴影或改变网格排列,但一定要保持高度宽阔的竖向布局。

#poster#opennana

觉得自己能猜中这条 prompt?来玩 Promptle