Galerie de prompts IA OpenZhenZhen · gpt-image-2.5 · image

Brune élégante sous un éclairage dramatique en clair-obscur

Fashionable Brunette in Dramatic Chiaroscuro Lighting

Brune élégante sous un éclairage dramatique en clair-obscur — découvrez cette création avec gpt-image-2.5 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Brune élégante sous un éclairage dramatique en clair-obscur — AI image creation
Brune élégante sous un éclairage dramatique en clair-obscur

Prompt

Prompt chinois

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。创作一张竖幅9:16时尚杂志特写,画面中只有一位明确成年的女性,拥有浓密蓬松的深棕色长发,面部身份与ref-10a-v1一致。她有榛棕色眼睛和利落的飞翼眼线,皮肤自然水润有纹理,颧骨带克制高光,嘴唇柔和微张,下颌线清晰。她以强烈、自信的直视镜头,呈现含蓄迷人的时尚编辑气质。她穿简洁的合身黑色T恤。采用平视角度,让她用一只手自然托起头顶发根,松散波浪发丝围绕并轻掠脸部;手部结构应自然。暖色硬光束斜切脸部与肩部,另一半落入深邃但保留细节的阴影。背景为中性灰,形成强烈明暗对照、暖金肤色高光与深黑阴影;眼睛清晰对焦,景深较浅。整体写实、优雅且具有杂志编辑质感,不要露骨化。不要额外人物、文字、水印、卡通处理、3D渲染感、模糊、扭曲解剖结构或多余手指。

Prompt anglais

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. Create a vertical 9:16 fashion-editorial close-up of one clearly adult woman with long, voluminous dark-brunette hair. Preserve her facial identity as ref-10a-v1. Give her hazel-brown eyes with sharp winged eyeliner, natural dewy skin texture, high cheekbones with a restrained highlight, softly parted lips and a defined jawline. Her expression is an intense, self-assured direct gaze with a subtly alluring editorial quality. She wears a simple fitted black T-shirt. Pose her at eye level, one hand naturally lifting the hair at the crown so loose waves frame part of her face; keep the hand anatomically natural. A hard beam of warm sunlight cuts diagonally across her face and shoulder, leaving the other half in deep but readable shadow. Use a neutral grey background, strong chiaroscuro contrast, warm golden skin highlights against deep blacks, sharp focus on the eyes and shallow depth of field. Keep the result photorealistic, elegant and magazine-editorial rather than explicit. No extra people, text, watermark, cartoon treatment, 3D-render look, blur, distorted anatomy or extra fingers.

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA