PROMPT RECORD · 公开作品详情
2x2 网格,16:9,为 4 部著名科幻电影执行此操作:{instructions: A = { $ SUBJECT 的轮廓与关键内部特征,…
公开 AI 图像创作记录，包含原图、提示词、使用模型、作者、标签与来源。
- 使用模型
- Nano Banana
- 创作者
- @Gdgtify
- 发布时间
- Sat Jul 11 03:54:00 +0000 2026
- 图片数量
- 1 张
完整提示词
源提示词按原始换行与标点完整保留2x2 grid, 16:9, do this for 4 famous scifi movies: {instructions: A = { silhouette and key internal features of $ SUBJECT, vectorized into black/white forms }
B = { deco ornamentation library: arches, stepped towers, vertical ribs, sunburst panels, circular medallions, chevrons, geometric grids }
C = { background metropolis: tall towers, bridges, arches, clock faces, abstract machinery }
D = { motion graphics: diagonal speed lines, fan rays, stepped echoes, theatrical silhouettes }
POSTER = (A ∩ subject_layer) ∪ (B ∩ ornamental_overlay) ∪ (C ∩ background_field) ∪ (D ∩ dynamic_layer) ∪ (ornamental_border)
Where:
- A is placed centrally, scaled to 70% of canvas.
- B elements are placed along the principal axes of A (symmetry axes, high‑curvature boundaries) by a space‑filling algorithm that respects a semi‑symmetrical grid.
- C fills the remaining canvas, its architectural shapes chosen to echo the angular frequency of A's silhouette (e.g., a curved subject spawns more arches, an angular subject spawns more stepped towers).
- D lines originate from the subject's action focal points, their direction and density derived from the pose vector.
- All elements strictly black, white, or limited 10% grey.}提示词译文
译文与源提示词分开显示，不替代原文2x2 网格,16:9,为 4 部著名科幻电影执行此操作:{instructions: A = { $ SUBJECT 的轮廓与关键内部特征,矢量化成黑白形态 }
B = { 装饰图案素材库:拱门、阶梯塔、垂直肋条、太阳放射板、圆形徽章、人字纹、几何网格 }
C = { 背景大都市:高塔、桥梁、拱门、钟面、抽象机械 }
D = { 动态图形:对角速度线、扇形放射、阶梯回声、戏剧化剪影 }
POSTER = (A ∩ subject_layer) ∪ (B ∩ ornamental_overlay) ∪ (C ∩ background_field) ∪ (D ∩ dynamic_layer) ∪ (ornamental_border)
其中:
- A 居中放置,缩放为画布的 70%。
- B 元素沿 A 的主轴(对称轴、高曲率边界)放置,通过遵循半对称网格的填充算法布置。
- C 填充画布的剩余空间,其建筑形态的选择旨在呼应 A 轮廓的角度频率(例如,曲线主体衍生更多拱门,棱角主体衍生更多阶梯塔)。
- D 线从主体的动作焦点发出,其方向与密度由姿态向量导出。
- 所有元素严格为黑、白,或限制在 10% 灰度。}