Which AI model should I use with this prompt?

This prompt is optimized for Sora, but works with any advanced text-to-image/video model that supports detailed scene generation.

Can I change the character or setting details?

Yes, the prompt includes customizable arguments for character description and scene setting—simply edit the text in the designated sections.

What exactly does the output include?

The output is a multi-section blueprint with a character turnaround, floor plans, elevations, a storyboard with camerawork notes, and lighting references.

GPT Image 2AI 提示词

Music Video Pre-Visualization Blueprint — Sora 2 Prompt

Generates a comprehensive music video pre-visualization blueprint with character references, set designs, storyboards, and lighting for a piano performance scene.

提示词 · EN

type

music video pre-visualization blueprint

header

title

{argument name="document title" default="MVピアノ演奏シーンカット構成・カメラワーク設計図"}

subtitle

キャラクター設定: コスチュームデザイン図 (画像左) を厳密に準拠美術設定: 洋館サロン平面図・美術設定 (画像右) を厳密に準拠

character

description

{argument name="character description" default="young woman with short platinum blonde bob hair, wearing a sleek black turtleneck and black pants"}

setting

description

{argument name="scene setting" default="dark, ornate Western-style mansion salon with a grand piano, patterned rugs, and chandeliers"}

layout

sections

item 1

title

キャラクター参照 (COSTUME DESIGN DIAGRAM)

position

top-left

count

labels

item 1

FRONT

item 2

SIDE

item 3

BACK

item 4

FRONT CLOSE-UP

item 5

SIDE CLOSE-UP

item 6

BACK CLOSE-UP

description

Character turnaround and close-up reference shots on a plain gray background.

item 2

title

美術設定参照 (洋館サロン：ピアノ演奏シーン)

position

top-right

count

labels

item 1

平面図(俯瞰)

item 2

室内立面図 (左側壁面)

item 3

右側立面図 (右側壁面)

item 4

断面図 (A-A')

description

Top-down floor plan of the piano room with numbered camera positions, plus side elevations and a cross-section diagram.

item 3

title

カット構成ストーリーボード

position

middle

count

labels

item 1

カット1: サイドミドル → 背後へオービット (序盤)

item 2

カット2: ピアノ手元インサート

item 3

カット3: 斜めアップ (歌唱・感情表現)

description

Storyboard panels showing the {argument name="main action" default="playing the piano and singing emotionally"}. Each panel includes descriptive text and a small camera diagram below it.

item 4

title

カメラワークの流れ (概念図)

position

middle-right

count

labels

item 1

① サイドミドル (開始)

item 2

② 背後へオービット完了

item 3

③ 手元インサート

item 4

④ 斜めアップ (歌唱)

description

Conceptual diagrams showing camera movement relative to the piano and singer.

item 5

title

照明イメージ (参考)

position

bottom-left

count

labels

item 1

自然光 (ステンドグラス越し)

item 2

シャンデリアのきらめき

item 3

壁付け燭台のあたたかな光

item 4

テーブルランプの柔らかな光

description

Lighting reference thumbnails showing different light sources in the room.

item 6

title

Direction and Tone Notes

position

bottom-right

count

labels

item 1

演出ポイント

item 2

推奨レンズ・トーン

description

Text boxes detailing direction points and recommended lens/tone settings.

关于这个提示词

Generates a comprehensive music video pre-visualization blueprint with character references, set designs, storyboards, and lighting for a piano performance scene. 它更适合作为 GPT Image 2 的概念艺术起点：先保留画面结构、主体关系和风格约束，再替换成你的品牌、人物或场景。

这条提示词使用结构化 JSON 组织信息，包含 type、header、character、setting、layout 等字段。保留这种层级能让模型更清楚地区分画面主题、布局、界面元素和细节约束。

使用时建议先小幅修改主体、场景、镜头和色调，再生成多个版本对比构图与细节。这样页面内容对用户有实际帮助，也避免把模型名或标签机械堆在正文里。