チャットGPT
どんでん返し — それは計画ではなかった
チャットGPT 向けに書かれました。結果と参照写真のサポートはモデルによって異なる場合があります。
このプロンプトを使う前に
リストされているツール: チャットGPT。完全な説明: 1,537 文字。 付属のサンプル画像 1 枚。
This prompt mentions reference images. Read its instructions for the required number of photos and identity constraints. We have not independently tested this prompt. Tool labels and supplied images are not a guarantee of compatibility or output quality.
プロンプト全文
Use the uploaded reference photo of the woman as the identity reference. Preserve her facial features, face shape, skin tone, eyes, nose, lips, and overall identity accurately.
Create a vertically split editorial composition inspired by the reference image.
TOP PANEL:
A photorealistic close-up portrait of the same woman, now wearing an oversized cream/off-white knitted sweater. Give her voluminous, naturally messy, wavy dark-brown hair with loose strands flying around her face. She has a surprised, slightly shocked expression, mouth gently open, eyebrows subtly raised, and her eyes looking toward the side. Warm soft indoor lighting, beige neutral background, realistic skin texture, cinematic photography, natural imperfections, shallow depth of field, high-end editorial photography.
BOTTOM PANEL:
Turn the exact same woman, pose, facial expression, hairstyle, and clothing into a detailed black-and-white hand-drawn graphite/ink sketch. Preserve her recognizable facial structure and expression. Use expressive rough pencil strokes, dense cross-hatching, loose sketch lines, textured off-white paper, artistic imperfections, and a fashion-editorial illustration aesthetic.
Add elegant typography on the left side of the lower panel:
“No.03”
“Plot Twist”
“—”
“That was not the plan.”
Keep the composition clean, minimal, sophisticated, and visually consistent between the photorealistic top panel and illustrated bottom panel. Vertical 4:5 format, high detail, realistic identity preservation, no extra people, no distortedプロンプト全体をコピーし、必要に応じて独自の参考写真を使用してください。
このプロンプトで提供されるサンプル画像。このプロンプトは当社のサイトではテストされていません。
