OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

次元の壁を破る:超写実的な善逸が本を破って飛び出す

Tearing the Dimensional Wall: Hyper-Realistic Zenitsu Bursts Out of the Book

「次元の壁を破る:超写実的な善逸が本を破って飛び出す」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

次元の壁を破る:超写実的な善逸が本を破って飛び出す — AI image creation
次元の壁を破る:超写実的な善逸が本を破って飛び出す

プロンプト

中国語プロンプト

创建一张令人惊叹的超写实8K照片:来自《鬼灭之刃》的我妻善逸正从一本巨大的实体日文漫画书中跨步走出。主要可见角色保持上传参考身份,拥有乱蓬蓬的金橙色头发、质感真实的鬼杀队制服、带白色三角纹的标志性黄色羽织、坚定锐利的眼神,并紧握带刀鞘的日轮刀。善逸一半身体站在略显磨损的木质卧室地板上,另一半仍留在书页中。厚重的日文漫画摊开在地板上;他跨出的边界处纸页猛烈撕裂、卷曲爆开,纸屑和墨粉飞舞。书页内部形成真实的黑白日漫画中画:粗糙黑色墨线、丰富网点阴影、夸张速度线、严格的日式从右到左阅读顺序,以及超现实四格漫画跨页。一个穿着同样服装、被困在格子里的漫画版善逸流着面条泪,惊慌地敲打格子边框,并恐惧地看着现实版自己撕裂画面。准确加入日文文字:出られない!、えええ!?、ドンドンドン。卧室背景柔焦,可见床脚和书架;自然午后侧窗光照亮撕裂纸张边缘与衣物织物纹理,强化现实和二维世界的对比。保持解剖结构清晰、纸张层次可信、接触阴影自然、电影感景深;避免难以辨认的额外文字、多余肢体、变形手部、污浊墨色和随机现代字体。

英語プロンプト

Create an astonishing hyper-realistic 8K photograph of Zenitsu Agatsuma from Demon Slayer stepping out of a giant physical Japanese manga book. Show the main visible character with the uploaded reference identity, messy gold-and-orange hair, a realistically textured Demon Slayer Corps uniform, iconic yellow haori with white triangle pattern, determined sharp eyes, and one hand tightly gripping a sheathed Nichirin blade. Half of his body stands on a slightly worn wooden bedroom floor while the other half remains inside the pages. The thick manga is spread open; paper tears violently around the boundary where he emerges, with curled pages, flying paper scraps and toner. Inside the book, show an authentic black-and-white Japanese manga picture-in-picture: rough black ink, rich screentone shading, exaggerated speed lines, strict right-to-left reading order and a surreal four-panel yonkoma spread. A trapped manga-style Zenitsu in matching clothes cries noodle tears, frantically knocks on the panel borders, and watches the real-world version tear through in terror. Include large Japanese sound/text marks exactly: 出られない!, えええ!?, ドンドンドン. Keep the bedroom softly out of focus with bed legs and bookshelves; natural afternoon side-window light catches the torn paper edges and fabric texture, emphasizing the contrast between reality and the two-dimensional world. Preserve clean anatomy, readable paper layers, convincing contact shadows and cinematic depth; avoid illegible extra text, extra limbs, distorted hands, muddy ink, or random modern typography.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成