【生成AI】画像生成の構図比べてみた|基本構図一覧【プロンプト無料公開】Midjourney / imageFX / ChatGPT / Whisk / imagen4 / Gemini / Sora / nano-banana / 生成AI / AI副業
こんにちはあいすけです
今回が画像生成における構図の種類を紹介したいと思います
「構図って何があってそれがどう映るのかわかんない、、、」
って感じたことはありませんか?
そこで今回は基本的な構図を紹介したいと思います。
プロンプトは僕がおすすめJSON形式のプロンプトで"camera"タグを変更するだけです!
ベース画像・プロンプト

{
"subject": "A detailed anime-style portrait of a beautiful 20-year-old Japanese woman with a mature and elegant allure. She has long, glossy dark brown hair flowing smoothly down her back with a gentle wave. Her large brown eyes soften warmly as she smiles, reflecting calm affection and reassurance. The corners of her lips lift into a gentle, natural smile that enhances her refined femininity without exaggeration. Her facial features remain smooth and youthful, with no visible wrinkles, and her expression conveys quiet warmth and emotional closeness. She wears tasteful casual private clothing—a fitted blouse and a simple skirt—that subtly accentuate her figure while maintaining elegance. Her posture is relaxed yet graceful, suggesting comfort in her private space.",
"camera": "medium shot, eye-level angle, frontal composition",
"background": "the interior of a quiet private home, softly furnished with a warm, intimate atmosphere that complements her gentle smile",
"style": "highly detailed anime illustration, vibrant yet balanced colors, clean lineart, soft shading, luxurious and modern anime aesthetic, similar to anything-v4/v5",
"lighting": "Soft, warm indoor lighting that gently highlights her face and hair, enhancing the warmth of her smile without harsh shadows",
"quality": "ultra high resolution, crisp outlines, vivid tones, refined details, strong emphasis on expression and character presence."
}このプロンプトの"camera"タグを変えるだけで構図が変わります!JSONプロンプトのいいところですね
正面構図(フロントビュー)
最も基本・強い訴求力


"camera": "front-facing composition, eye-level camera angle, subject centered, direct view"三分割構図(ルール・オブ・サード)
自然で洗練された印象


"camera": "rule of thirds composition, subject positioned on left third, eye-level angle, balanced negative space"ローアングル構図(見上げ)
支配・自信・ヒーロー感


"camera": "low-angle shot, camera positioned below subject, upward perspective, heroic framing"ハイアングル構図(見下ろし)
弱さ・内省・客観性


"camera": "high-angle shot, camera positioned above subject, downward perspective, observational framing"
横顔構図(プロファイル)
知的・静的・美しい


"camera": "side profile composition, subject facing sideways, eye-level angle, clean profile framing"背中構図(後ろ姿)
物語性が最も強い


"camera": "rear-view composition, subject seen from behind, centered framing, cinematic depth"
対角線構図(動き重視)
エネルギー・躍動感


"camera": "diagonal composition, subject aligned along diagonal axis, dynamic perspective"ロングショット(Long Shot)
主役+世界観を同時に見せる


"camera": "long shot, full-body framing, eye-level angle, subject fully visible within the frame"
いかかがでしたか?
皆さんもぜひこのプロンプトをもとに遊んでみてください!
今回紹介した画像以外にも、
これまでに自分で生成したAI画像をギャラリーとしてまとめています。
よければ、他の作品もチェックしてみてください。
プロンプト作成の仕方は↓
