OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

一貫性のある四コマ映画風絵コンテデザイン

Coherent Four-Panel Cinematic Storyboard Design

「一貫性のある四コマ映画風絵コンテデザイン」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

一貫性のある四コマ映画風絵コンテデザイン — AI image creation
一貫性のある四コマ映画風絵コンテデザイン

プロンプト

中国語プロンプト

创作一张连贯的电影感四格分镜图,采用干净的2×2网格,四格尺寸相同、间距窄且均匀。以a small folded paper boat drifting from a rain-soaked city canal toward a quiet dawn作为短篇叙事。明确呈现四个依次推进的节点:(1)交代环境并引入叙事母题;(2)让主体遇到一个小障碍;(3)展示障碍发生变化时的转折点;(4)以安静的最终画面收束动作。各格保持主体设计、尺度、运动方向、地理关系和重复地标一致,让画面读作同一连续事件,而不是四张互不相关的图。使用consistent low water-level wide shots with one recurring stone bridge landmark,并通过cool slate-blue rain shifting gradually to muted peach and warm dawn gold逐步改变氛围。采用电影化构图,前景与背景层次清楚,空间氛围克制,自然反射且光线连续。不要人物、文字、字幕、分格标题、数字、标志、水印、对话气泡或额外嵌套小画格。纸船旅程是因来源正文没有故事内容而补入的可替换示例默认值,并非声称它就是来源原故事。

英語プロンプト

Create one cohesive cinematic four-panel storyboard as a clean 2×2 grid, with four equally sized panels and narrow even gutters. Use a small folded paper boat drifting from a rain-soaked city canal toward a quiet dawn as the short narrative. Show four unmistakably sequential beats: (1) establish the setting and introduce the motif, (2) let it move into a small obstacle, (3) show the turning point as the obstacle changes, and (4) resolve the motion in a quiet final image. Keep the same subject design, scale, direction of travel, geography, and recurring landmark from panel to panel so the sequence reads as one continuous event rather than four unrelated images. Use consistent low water-level wide shots with one recurring stone bridge landmark and progress the atmosphere through cool slate-blue rain shifting gradually to muted peach and warm dawn gold. Cinematic composition, carefully staged foreground and background, restrained atmospheric depth, natural reflections and consistent lighting continuity. No people, lettering, captions, panel labels, numbers, logos, watermark, speech bubbles, or extra inset frames. The paper-boat journey is a replaceable sample default added because the source record contains no story text; it is not asserted to be the original source narrative.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成