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

窓辺の自然光で描く写実的な東アジア女性の肖像

Realistic East Asian Woman Portrait by Window in Natural Light

「窓辺の自然光で描く写実的な東アジア女性の肖像」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。

窓辺の自然光で描く写実的な東アジア女性の肖像 — AI image creation
窓辺の自然光で描く写実的な東アジア女性の肖像

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。 生成一张竖构图、写实自然的人像照片。画面主体是一位明确成年的东亚女性,采用身形以上的近景构图,站在简洁的白色肌理墙与深色窗框旁。人物身体略微侧向窗户,一侧肩膀自然向前,视线望向窗外,不看镜头。保持精致但真实的椭圆脸型、自然眼型、鼻唇比例与成年感;深色中长层次发,轻薄碎刘海,少量发丝自然落在额头和脸侧,佩戴小巧银色耳圈,穿简约黑色上衣,妆容清淡克制。窗外自然光从侧面照亮面部,在另一侧脸颊形成柔和阴影。呈现真实毛孔、细小皮肤纹理、唇部纹理、细碎发丝和轻微左右不对称,具有50毫米纪实人像镜头的自然空间感。背景保持干净、简单、无杂物。禁止塑料皮肤、过度磨皮、夸张大眼、网红滤镜、影棚式强光、额外首饰、文字、商标和水印

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. Generate a vertical composition, realistic and natural portrait photo. The main subject is a clearly adult East Asian woman, using a close-up composition from the chest up, standing next to a simple white textured wall and a dark window frame. The person's body is slightly turned towards the window, one shoulder naturally forward, gaze looking out the window, not looking at the camera. Maintain an exquisite but realistic oval face shape, natural eye shape, nose-to-lip ratio, and sense of adulthood; dark medium-length layered hair, light wispy bangs, a few strands of hair naturally falling on the forehead and side of the face, wearing small silver hoop earrings, wearing a simple black top, light and restrained makeup. Natural light from the window illuminates the face from the side, forming a soft shadow on the other cheek. Present real pores, fine skin texture, lip texture, fine strands of hair, and slight left-right asymmetry, with the natural sense of space of a 50mm documentary portrait lens. The background remains clean, simple, and clutter-free. Prohibit plastic skin, excessive skin smoothing, exaggerated large eyes, influencer filters, studio-style strong light, extra jewelry, text, logos, and watermarks.

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