OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image

Concert signing snapshot by the red railing

Concert signing snapshot by the red railing — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.

Concert signing snapshot by the red railing — AI image creation
Concert signing snapshot by the red railing

Prompt

Chinese prompt

创作一张室内演唱会签售区域的真实全身抓拍,竖幅4:5。画面仅有一位明确成年的女性,其面部身份、发型与外貌仅依据已附加的角色参考图 ref-10a-v1。她站在红色人群护栏旁,平静地在放于护栏表面的空白签名卡上签字。她穿着得体、全内衬的soft pink蕾丝风连衣裙,搭配轻便运动夹克与无品牌小肩包;覆盖自然舒适,不使用透视面料。明亮的场馆灯光从上方发亮,背景为昏暗柔焦的场馆,呈现空座位、彩色散景与模糊舞台灯光,以环境传达活动氛围,不出现其他人物。自然机顶闪光、真实皮肤与织物质感、随手手机快照构图,环境光与闪光形成柔和反差。签名卡保持空白,或只带不可辨认的抽象笔迹。不得出现观众、粉丝、伸入画面的手、可读文字、标志、水印、额外人物,也不得模仿具名名人。必须附加角色参考图;来源肖像并未确认为本地已附加素材。

English prompt

Create a realistic candid full-body 4:5 snapshot at an indoor concert signing area. Show only one clearly adult woman, with facial identity, hairstyle and appearance based solely on the attached character reference ref-10a-v1. She stands beside a red crowd barrier, calmly signing a blank autograph card resting on the barrier surface. She wears a tasteful fully lined soft pink lace-inspired dress with a light sporty jacket and a small unbranded shoulder bag; natural, comfortable coverage and no sheer fabric. Bright arena lights glow overhead; the background is a dark, softly blurred venue with empty seating, colored bokeh and indistinct stage light, conveying event energy without showing other people. Natural on-camera flash, realistic skin and fabric texture, spontaneous smartphone-snapshot framing, gentle ambient/flash contrast. Keep the autograph card blank or marked only with non-legible abstract strokes. No audience, fans, hands entering frame, readable text, logos, watermark, extra people or imitation of a named celebrity. The character reference is required and must be attached; the source portrait is not a verified local attachment.

View API model

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas