Galerie de prompts IA OpenZhenZhen · gpt-image-2.5 · image

Samouraï 2D dans le métro de Tokyo

2D Samurai in the Tokyo Subway

Samouraï 2D dans le métro de Tokyo — découvrez cette création avec gpt-image-2.5 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Samouraï 2D dans le métro de Tokyo — AI image creation
Samouraï 2D dans le métro de Tokyo

Prompt

Prompt chinois

创作一张东京拥挤地铁车厢的超写实电影感照片。车厢与乘客保持真实摄影质感;在一张真实地铁座位上坐着一名疲惫的成年武士角色,但他始终是清晰的二维黑白漫画绘画:干净有力的墨线、细密交叉排线和纸面笔触,不要变成真人或立体CG。角色的面部身份绑定固定参考 ref-10a-v1,参考只约束脸部身份,不改变二维墨绘表现。车厢灯光像照在真实物体上一样与画中角色发生光影互动:在墨线和排线上出现克制、可信的高光与明暗,但保持平面漫画轮廓。武士一只手握着一罐看起来真实拍摄的无品牌汽水罐,罐子有金属反光并自然伸出二维画面,与真实车厢空间产生轻微的二维/三维交界效果。用中景构图,让拥挤乘客、扶手与车门构成背景层次,同时武士和汽水罐清楚可辨;避免把路人画成主要人物。细节清晰、电影感构图,不承诺特定输出分辨率。无可读商标、额外文字、水印或乱码。

Prompt anglais

Create a hyper-realistic cinematic photograph of a crowded Tokyo subway car. Keep the car and surrounding commuters photographic. Seated on a real train seat is a tired adult samurai character rendered unmistakably as a flat black-and-white 2D manga drawing: clean confident ink contours, fine cross-hatching and visible paper-like linework; never turn the character into a live-action person or 3D CGI. Bind the character's facial identity to the fixed reference ref-10a-v1, using it only for identity while preserving the 2D ink-drawn treatment. Let the subway lighting interact plausibly with the drawing as if falling across the printed figure, creating restrained highlights and shadows along the ink strokes and hatching while preserving its flat illustrated outline. The samurai holds a realistic photographic, unbranded soda can in one hand; give the can metallic reflections and let it extend naturally across the 2D/3D boundary into the real train space. Use a medium composition with crowded passengers, handrails and doors establishing depth, while keeping the samurai and can legible; commuters remain background figures. Render crisp fine detail and cinematic composition without promising a specific output resolution. No readable logos, added text, watermark or garbled lettering.

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA