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

巨大広告板に映る同じ都市の女性

The Same Urban Woman on a Giant Billboard

「巨大広告板に映る同じ都市の女性」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。

巨大広告板に映る同じ都市の女性 — AI image creation
巨大広告板に映る同じ都市の女性

プロンプト

中国語プロンプト

创作一张竖版9:16电影感都市夜景写实照片。现代街角前景中,一位约25岁、明确成年的时尚女性全身入镜,面向镜头并带着微妙自信的笑容,一只手指向她身后的高层玻璃建筑。她的长直棕发与面部身份忠实匹配固定参考ref-10a-v1。她穿浅粉色T恤、敞开的黑色休闲外套、修身深色牛仔裤和擦亮的黑色皮鞋。身后的玻璃楼体上设置一块巨型清晰数字广告牌,展示同一位女性身着同一套服装的高级时装杂志封面肖像;广告牌中的脸也必须与ref-10a-v1身份一致。这只是屏幕里的第二次呈现,不是在街上新增另一位人物。广告牌肖像可采用从容的杂志封面姿势。画面中唯一醒目可读标题必须准确写作“沃特风格”(默认:沃特风格);标题下只可有刻意模糊、无法辨认的微小杂志填充字,不得出现其他可读文案。周围是充满活力的现代夜间城市:玻璃建筑、没有可读文案的发光霓虹招牌、带克制运动模糊的行驶车辆、亮起的广告屏和映照灯光的雨后湿路面。使用电光青与洋红霓虹,辅以温暖琥珀色(默认:电光青与洋红霓虹,辅以温暖琥珀色),以混合霓虹环境光和广告牌定向光照明,阴影有情绪但细节仍清楚;呈现真实湿地反光和35mm摄影透视。让女性、指向动作、对应的广告牌肖像和玻璃楼体在同一画面中一目了然,以浅景深和柔和城市散景收束背景。保留自然的成年面部特征与真实摄影肤质;不要额外的前景行人、名人相似脸、无关文字、其他醒目标志或水印。使用工具支持的最高图像细节,不承诺工具不支持的精确8K像素分辨率。

英語プロンプト

Create a cinematic, photorealistic vertical 9:16 urban-night photograph. At a modern city street corner, show ONE clearly adult stylish woman, about 25, standing full-length in the foreground facing the camera with a subtle confident smile; she points with one hand toward the tall glass building directly behind her. Her long straight brown hair and facial identity must faithfully match fixed reference ref-10a-v1. Dress her in a soft-pink T-shirt under an open black casual jacket, fitted dark jeans and polished black shoes. On the building behind her, place one enormous crisp digital billboard showing a high-fashion editorial portrait of THE SAME woman in the same outfit, with the same ref-10a-v1 facial identity; this is a second depiction on the screen, not another person standing in the street. The billboard portrait may use a poised magazine-cover stance. Its only large readable headline must be exactly “VOUGHT STYLE” (default: VOUGHT STYLE); include only tiny deliberately indistinct magazine-style filler beneath it, with no other legible wording. Surround the scene with a vibrant modern night city: glass architecture, luminous neon signs without readable copy, moving cars with restrained motion blur, glowing billboards and rain-wet pavement reflecting the lights. Use electric cyan and magenta neon with warm amber accents (default: electric cyan and magenta neon with warm amber accents), mixed neon ambient light and directional billboard light, moody but detailed shadows, realistic wet reflections and a 35mm photographic perspective. Keep the woman, her pointing gesture, the specific billboard portrait and the glass building all clearly readable in one unified frame, with shallow depth of field and soft city-light bokeh. Preserve natural adult facial features and realistic photographic skin; no extra foreground people, no celebrity likeness, no unrelated text, no additional prominent logos or watermark. Use the highest supported image detail without promising an unsupported exact 8K pixel resolution.

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