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

雨の夜の街の冷徹なギャング ネオノワール写実的肖像

Rainy Night Street Cold Gangster Neo-Noir Realistic Portrait

「雨の夜の街の冷徹なギャング ネオノワール写実的肖像」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。

雨の夜の街の冷徹なギャング ネオノワール写実的肖像 — AI image creation
雨の夜の街の冷徹なギャング ネオノワール写実的肖像

プロンプト

中国語プロンプト

创建一幅3:4竖幅、超电影感的照片级写实半身人像:使用上传的面部参考,以中等身份强度和较高影响力保持整体面貌,同时允许自然的轻微变化,只进行细微风格化,不要僵硬锁定脸部。表现一名年轻成年男子在1920年代城市夜街上缓慢而自信地向前走。他一只手拿着点燃的香烟,另一只手放在大衣口袋里;肩膀平直但放松,头部略微倾斜,直视镜头,表情冷峻、无畏且具有权威感。准确穿着1920年代街头黑帮服装:深色厚重羊毛长大衣,布料略微湿润,内搭深色扣好的三件套西装和马甲,领带松开,低压平顶帽。场景包含湿润鹅卵石路、复古街灯、砖砌建筑和老式店面;远处人物虚化为雨夜灯光散景,香烟烟雾与雨丝交融。以暖色街灯为主光、冷色月光为辅光,采用高反差新黑色电影调色:深沉阴影、发光琥珀高光、冷蓝灰环境,并突出面部、眼睛、雨滴和大衣纹理。使用平视机位、85mm镜头、f/1.4超浅景深,眼睛清晰对焦,具有电影定格感和细微胶片颗粒。表现真实皮肤、羊毛和湿润反射的细节;排除卡通、动漫、插画、绘画感、塑料皮肤、脸部扭曲、错误解剖、多余肢体、低分辨率、过曝、现代物件或汽车、标志和文字。

英語プロンプト

Create a vertical 3:4, ultra-cinematic photorealistic half-body portrait of a young adult man using the uploaded face reference with medium identity strength and high influence: preserve the overall look, allow minor natural variation and only subtle stylization; do not lock the face rigidly. Show him walking forward in a slow, confident stride on a 1920s city street at night. He holds a lit cigarette in one hand and keeps the other inside his overcoat pocket; shoulders square but relaxed, head slightly tilted, eyes looking directly at the camera with a cold, fearless, authoritative expression. Dress him accurately as a 1920s street gangster: long dark heavy-wool overcoat with slightly wet fabric, dark buttoned three-piece suit and waistcoat, loosened tie, and a flat cap worn low. Set the scene among wet cobblestones, vintage street lamps, brick buildings and old shopfronts, with distant figures blurred into rainy light bokeh and cigarette smoke mixing with the rain. Use warm street lamps as the key light and cool moonlight as fill, with high-contrast neo-noir grading, deep shadows, glowing amber highlights, cool blue-gray surroundings, and emphasis on the face, eyes, raindrops and coat texture. Photograph at eye level with an 85mm lens, f/1.4 ultra-shallow depth of field, eyes in focus, cinematic freeze and subtle film grain. Render detailed realistic skin, wool and wet reflections; exclude cartoon, anime, illustration, painterly style, plastic skin, warped face, bad anatomy, extra limbs, low resolution, overexposure, modern objects or cars, logos and text.

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