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

陽光の窓辺で読書と蝶の牧歌的フィルムポートレート

Pastoral Film Portrait of Reading by a Sunny Window with Butterflies

「陽光の窓辺で読書と蝶の牧歌的フィルムポートレート」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。

陽光の窓辺で読書と蝶の牧歌的フィルムポートレート — AI image creation
陽光の窓辺で読書と蝶の牧歌的フィルムポートレート

プロンプト

中国語プロンプト

以用户上传的参考图作为主要可见人物的面部身份依据。在不与身份参考冲突的情况下,保留来源的场景、造型、服装、姿势、构图、光线和视觉风格。仅使用人像参考其面部身份、发型、肤色和自然气质;重新创作构图,不要照搬参考照片。创作一张梦幻写实的竖版4:5胶片感肖像:一位明确成年的女性坐在洒满阳光的大窗边,手捧翻开的书安静阅读。她穿柔软的白色夏装与精致蕾丝裙,搭配长袜,身旁放着一只小草编包。几只自然可信的蝴蝶轻轻停在她发间和肩侧,不遮挡面部。窗外可见郁郁葱葱的绿色花园。温暖逆光透过薄透白窗帘,窗帘轻柔飘动;画面保留柔和阴影和自然肌肤纹理。融合电影感田园氛围、克制的35毫米胶片颗粒与随手手机拍摄的真实感。面部清晰对焦,避免浓重修饰,并保持手部与蝴蝶形态可信。负面约束:不要水印、标志、文字、AI感伪影、动漫、卡通、洋娃娃脸、塑料或过度磨皮的皮肤、虚假光影、扭曲手部、多余手指、错误的人体结构、模糊面部、浓妆、不真实的蝴蝶、低画质或过度饱和。

英語プロンプト

Use the uploaded reference image as the facial identity reference for the main visible person. Keep the source scene, styling, clothing, pose, composition, lighting and visual style unless they conflict with the identity reference. Use the portrait only for the woman's facial identity, hairstyle, skin tone, and natural presence; create a new composition rather than copying the reference photograph. Create a dreamy, realistic 4:5 film-style portrait of one clearly adult woman seated beside a large sunlit window, holding an open book and reading quietly. Dress her in a soft white summer outfit with a delicate lace skirt, long socks, and a small straw bag nearby. Let a few natural-looking butterflies rest gently around her hair and shoulder without obscuring her face. Outside the window, show a lush green garden. A sheer white curtain moves softly in the warm backlight; use gentle shadows and natural skin texture. Blend a cinematic cottagecore atmosphere with subtle 35 mm film grain and the casual realism of a phone photograph. Keep the face in focus, avoid heavy retouching, and preserve believable hands and butterflies. Negative constraints: no watermark, logo, text, AI-looking artifacts, anime, cartoon, doll-like face, plastic or over-smoothed skin, fake lighting, distorted hands, extra fingers, bad anatomy, blurry face, harsh makeup, unrealistic butterflies, low quality, or oversaturated colors.

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