空間レイヤーに浮かぶ立体透視タイポグラフィ
30

GPT Image 2.0 プロンプト - 空間レイヤーに浮かぶ立体透視タイポグラフィ

作成者
xie underwood
モデル
GPT Image 2.0
アスペクト比
2:3
画質
1K

プロンプト

TITLE: [promptsref] PROJECT: [brand / exhibition / campaign] MEDIUM: [threads / points / volumetric light / moiré layers / other] SPATIAL MECHANISM: [alignment / depth / interference / optical reveal] SUBJECT: [small human figure / none] MAIN COLOR: [accent color] FORMAT: 9:16 vertical poster Create a contemporary volumetric perspective typography poster where the main word exists through elements distributed across real three-dimensional space rather than on a flat surface. Build the word from the chosen medium: suspended threads, physical points, beams of light, layered line grids, or another spatial system. Individual elements should appear fragmented or abstract from other viewpoints, but align into one bold, readable word from the intended camera position. Keep clear foreground, midground and background depth. The typography should dominate roughly 35–55% of the composition while still revealing how the spatial structure creates it. For threads, build each letter from lines at different depths. For points, use suspended objects to form volumetric letter shapes. For light, let separate beams visually combine into the word. For moiré, use overlapping line layers and interference to reveal the typography. If a person is included, keep them small and use them mainly as a scale reference, with realistic occlusion between the subject and the spatial elements. Use one strong accent color with a restrained architectural palette. Add only minimal editorial information such as project name, location, date and one short thematic line. Overall style: perspective typography, volumetric installation, spatial graphic design, contemporary exhibition art, clean hierarchy and strong visual impact. Avoid flat printed text, ordinary 3D letters, weak depth, random distortion, neon sci-fi styling, excessive filler copy, cluttered backgrounds, watermarks and numbering.

この画像プロンプトを使う

もっと見る
  1. 01

    プロンプトから始める

    プロンプトをコピーするか、生成ボタンで開きます。

  2. 02

    自分のアイデアを加える

    被写体や背景、細部を調整し、必要に応じて参考素材を追加します。

  3. 03

    生成前に確認する

    モデルと出力設定を確認してください。生成結果は毎回異なる場合があります。

似ている作品

次のアイデアを探していますか?

プロンプトを検索するか、画像からプロンプトを作成しましょう。