原文
GPT Image 2 · en
Analyze the uploaded image and preserve the original subject, composition, and lighting. Do not alter the identity or structure of the main subject. Add playful, {argument name="doodle style" default="hand-drawn doodles"} that interact directly with the subject in the image. The doodles should mimic, follow, or exaggerate the shapes, gestures, or motion present-such as outlining poses, extending limbs, adding motion lines, or creating imaginative elements that
"respond" to the subject.
Ensure the doodles feel naturally integrated into the scene, as if they were drawn on top of the photo with intention.
Use a sketchy, imperfect, hand-drawn style with organic lines, slightly uneven strokes, and a casual illustrated feel.
Include whimsical handwritten text elements placed around the image. The text should match the mood or implied context of the scene, with a playful and spontaneous tone.
Avoid fixed phrases-generate context-aware, creative, and humorous text that fits each unique image. Maintain a balanced composition so the doodles enhance the image without overwhelming the original subject. Keep the overall aesthetic fun, expressive, and social-media-ready. High resolution, clean overlay, vibrant yet natural color harmony.使い方
- パラメータを含む元のプロンプト全体をコピーします。
- 被写体、文字、構図を自分の要件に置き換えます。
- 対応する外部ツールで使用し、利用条件と結果を確認します。
開始前の確認
- 参考画像の要否を確認してください。互換モデルの表示は、そのモデルでの生成実績を意味しません。
- 名前、数値、画像内の文字を確認します。モデルの版で結果は変わります。
