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

Selfie fictif des coulisses d'un concert

Fictional backstage concert selfie

Selfie fictif des coulisses d'un concert — découvrez cette création avec gpt-image-2.5 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Selfie fictif des coulisses d'un concert — AI image creation
Selfie fictif des coulisses d'un concert

Prompt

Prompt chinois

将 identity_reference(实际固定照片中的同一位成年女性)仅用于保持其可辨识的面部、发型和比例。创作一张新的虚构、非纪实 3:4 抓拍人像,呈现类似 iPhone 前置摄像头的视觉质感:她走在现场演唱会后台走廊中,微微转身,带着放松而疲惫的演后浅笑和轻柔笑声;头发凌乱、像被风吹动,神情自然未经摆拍。不要暗示参考照片中的女性现实中参加或演出了某场演唱会。让画面有偶然抓拍感:手臂伸出、镜头看似靠得过近、倾斜的偏心裁切,以及头发和手臂的轻微动态模糊。使用弱光噪点、舞台灯光导致的轻微过曝、镜头光晕、轻微对焦不一致和自然手机摄像头处理。后台走廊柔和虚化,隐约呈现 LED 灯、设备箱、远处工作人员剪影和场馆光芒;不要额外前景人物。保留真实皮肤纹理和同一成年身份,不做美颜修图或棚拍构图;避免不同面孔、变形手部、水印或虚构文字。iPhone/前置摄像头仅描述视觉风格,不声称实际设备、EXIF、模型或输出格式发生变化。

Prompt anglais

Use identity_reference, the actual fixed photo of the same adult woman, only to preserve her recognizable face, hairstyle and proportions. Create a new fictional, non-documentary 3:4 candid portrait with an iPhone-like front-camera visual aesthetic: she is walking through a live-concert backstage corridor, turning slightly with a relaxed, tired post-performance smile and soft laugh; her hair is messy and windblown, and her expression is natural and unposed. Do not imply that the reference woman actually attended or performed at a concert. Make the framing feel accidental, with an extended arm, camera apparently too close, tilted off-center crop, and slight motion blur on hair and arm. Use low-light grain, mild stage-light overexposure, lens bloom, slight focus inconsistency, and natural phone-camera processing. Keep the hallway softly blurred while suggesting LED lights, equipment cases, distant staff silhouettes and arena glow; no extra foreground people. Preserve realistic skin texture and the same adult identity, without beauty retouching, studio composition, different face, distorted hands, watermark or invented text. The iPhone/front-camera wording describes visual style only; do not claim actual device, EXIF, model, or output-format changes.

Voir le modèle API

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