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

古風ファンタジー 夜の提灯人物撮影

Ancient Fantasy Night Lantern Portrait Photography

「古風ファンタジー 夜の提灯人物撮影」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。

古風ファンタジー 夜の提灯人物撮影 — AI image creation
古風ファンタジー 夜の提灯人物撮影

プロンプト

中国語プロンプト

仅使用上传人像作为明确成年的身份参考,保留真实五官、脸型、肤色、眼形、鼻唇结构、下颚线、自然面部不对称和可辨识发型。为同一人创作一张9:16竖幅的超写实电影感古风玄幻夜晚人像,从旧石墙旁的低角度近中景拍摄;主体微微俯在墙上,平静、神秘、优雅地向下看向相机。一只手自然提着发出温暖金光的传统纸灯笼,灯光柔和照亮脸部、手部、袖子和附近石块。服装为原创的中国古代灵感高级造型,结合柔软分层丝绸、精致刺绣、飘逸半透明袖子、适度纹理肩部细节和优雅配饰,风格精致、品味高雅,不直接复制任何历史服饰或现有角色。背景是深夜古老石建筑与淡雾,松散发丝被风自然吹动;后方有戏剧性边缘光,前景灯笼提供暖光,形成金色高光与深影的强烈对比。使用真实皮肤纹理、可见细微毛孔、可信手指与手腕、自然布料运动、电影级景深和高端电影剧照质感。灯笼位于下方前景,石墙形成引导线。避免现代服装、廉价戏服、复制套装、身份漂移、假脸、塑料皮肤、动漫/插画/卡通感、额外人物、手部畸形、过曝面部、随机文字、logo、水印和美颜滤镜。

英語プロンプト

Use the uploaded portrait only as the identity reference for the clearly adult subject, preserving real facial features, face shape, skin tone, eye shape, nose and lip structure, jawline, natural asymmetry, and recognizable hairstyle. Create a 9:16 vertical ultra-realistic cinematic ancient-inspired fantasy night portrait of the same person, shot from a low-angle close-medium view beside an old stone wall. The subject leans slightly over the wall and looks down toward the camera with a calm, mysterious, elegant expression. One hand naturally holds a traditional paper lantern emitting warm golden light that softly illuminates the face, hands, sleeves, and nearby stones. Dress the subject in an original premium ancient-Chinese-inspired outfit with layered silk, delicate embroidery, flowing translucent sleeves, restrained textured shoulder detail, and graceful accessories; keep it tasteful and rich without copying any historical costume or existing character. Set the scene against old stone architecture and faint mist at night, with loose hair moved naturally by wind, dramatic rim light from behind, and warm lantern light in the lower foreground creating strong contrast between gold highlights and deep shadows. Use realistic skin texture, subtle pores, believable fingers and wrists, natural fabric motion, cinematic depth of field, and high-end movie-still quality. Let the stone wall create leading lines. Avoid modern clothing, cheap costume styling, copied outfits, identity drift, fake face, plastic skin, anime/illustration/cartoon appearance, extra people, malformed hands, overexposed face, random text, logos, watermarks, or beauty filters.

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