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

渋谷のブルーアワーに撮る都市スナップ

City Street Photography at Shibuya's Blue Hour

「渋谷のブルーアワーに撮る都市スナップ」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。

渋谷のブルーアワーに撮る都市スナップ — AI image creation
渋谷のブルーアワーに撮る都市スナップ

プロンプト

中国語プロンプト

创作一张9:16竖幅写实街头时尚肖像,场景为黄昏蓝调时刻的东京涩谷十字路口。一位明确成年的日本女性(约22岁)依据已配置身份参考图呈现,保留可辨认的面部身份;她有深棕近黑色中长波波头与齐刘海,神情安静、略带忧郁,目光低垂。全身入画,人物站在人行道靠近路缘处,姿态自然放松,一只手插在口袋里,肩部微收。她穿红色与米色拼接的棒球夹克,白色条纹罗纹边,内搭纯白T恤;厚实芥末黄色针织围巾带清晰流苏,搭配褪色宽松阔腿蓝色牛仔裤(膝部明显磨损和破洞)以及红黑篮球风运动鞋、黑鞋带。后方是涩谷路口霓虹与城市灯光形成的柔和散景,少量行人略带动态模糊;路面有温暖的傍晚反光,不下雨。使用35–50mm街拍视角、略浅景深,主体面部与服装清晰,保留自然皮肤和发丝质感,暖色高光与微冷阴影形成克制的电影调色。保持完整全身和鞋履,不裁切;背景招牌不可读,不新增文字、品牌标志、水印或边框。避免肢体畸形、额外手指或肢体、塑料磨皮、过曝及死黑。

英語プロンプト

Create a realistic 9:16 vertical street-fashion portrait at Tokyo's Shibuya Scramble Crossing during dusk blue hour. Depict a clearly adult Japanese woman, around 22, using the configured identity reference to preserve her recognizable facial identity. She has a dark-brown-to-black medium bob with straight bangs and a quiet, slightly melancholic expression with her gaze lowered. Show her full body standing naturally near the sidewalk curb, one hand in a pocket and shoulders subtly tucked in. Dress her in a red-and-beige varsity jacket with white striped ribbing over a plain white T-shirt, a thick mustard-yellow knitted scarf with visible fringe, faded relaxed wide-leg blue jeans with pronounced wear and large knee rips, and red-and-black basketball-style sneakers with black laces. Behind her, neon and city lights at the crossing form soft bokeh, with a few pedestrians gently motion-blurred. Warm evening light reflects on the pavement; no rain. Use a 35–50 mm street-photography viewpoint and moderately shallow depth of field, keeping the woman's face, outfit, full silhouette, and shoes clear. Preserve natural skin texture and individual hair strands; use restrained cinematic grading with warm highlights and slightly cool shadows. Do not crop her body. Keep background signage unreadable and add no extra text, brand marks, watermark, or border. Avoid malformed anatomy, extra fingers or limbs, plastic-looking skin, blown highlights, and crushed blacks.

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