OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

梦幻彩虹棱镜光漏电影感夜晚人像

Dreamy Rainbow Prism Light Leak Cinematic Night Portrait

A 9:16 hyper-realistic night portrait using an uploaded identity reference, experimental flash, rainbow prism streaks, windblown hair, and a dark seaside atmosphere.

梦幻彩虹棱镜光漏电影感夜晚人像 — AI image creation
梦幻彩虹棱镜光漏电影感夜晚人像

提示词 / Prompt

中文提示词

仅使用用户上传的肖像作为同一位成年人物的身份和发型参考。保持面部身份、脸型、五官、肤色、发型、发量和自然外观，但不要复制参考图中的服装、背景、光线、姿势或图像质量。在黑暗海边或开放户外夜景附近创作一张超写实电影感夜晚肖像。采用特写半身构图，主体越过一侧肩膀微微回头看向相机，表情冷静、柔和而神秘，面部清晰，保留自然皮肤纹理、真实眼睛、柔和唇形和细致妆容。使用实验性闪光摄影：水平彩虹棱镜光漏、横跨额头和上半脸的明亮白色镜头光晕，以及穿过画面的蓝、青、橙、红、绿光条纹。加入可信的长曝光运动模糊、胶片颗粒、柔光、轻微复古数码相机质感和被风轻轻吹动、带有受光发丝的头发。背景保持非常暗并略微虚化。主体穿浅色无袖连衣裙，具有柔软面料质感、细微荷叶边和自然肩线，造型得体且不暴露。氛围梦幻、电影化、神秘、高级时装而写实。垂直9:16，真实相机视角，自然解剖结构，高级社论肖像品质。不要文字、标志、水印、签名、AI标签、动漫、插画、娃娃脸、塑料或过度平滑的皮肤、改变身份、不同发型、解剖变形、多余手指、面部扭曲、面部模糊、低分辨率、混乱构图、面部高光溢出、廉价滤镜、随机物体、重复人物、不自然脖子或肩膀以及噪点伪影。

English prompt

Use the uploaded portrait only as the identity and hairstyle reference for the same adult person. Preserve facial identity, face shape, features, skin tone, hairstyle, hair volume, and natural appearance, but do not copy the source clothing, background, lighting, pose, or image quality. Create a hyper-realistic cinematic night portrait near a dark seaside or open outdoor environment. Use a close half-body composition as the subject turns slightly back toward the camera over one shoulder, with a calm, soft, mysterious expression, clear natural skin texture, realistic eyes, soft lips, and subtle makeup. Apply experimental flash photography with horizontal rainbow prism light leaks, a bright white lens flare across the forehead and upper face, and blue, cyan, orange, red, and green light streaks crossing the frame. Add believable long-exposure motion blur, film grain, soft glow, subtle vintage digital-camera texture, and gently windblown hair with illuminated strands. Keep the background very dark and slightly blurred. Dress the subject in a light-colored sleeveless dress with soft fabric texture, subtle ruffles, and a natural shoulder line, keeping it tasteful and non-revealing. Mood: dreamy, cinematic, mysterious, high-fashion, photorealistic. Vertical 9:16, realistic camera perspective, natural anatomy, premium editorial quality. No text, logos, watermarks, signatures, AI labels, anime, illustration, doll face, plastic or over-smoothed skin, changed identity, different hairstyle, deformed anatomy, extra fingers, distorted face, blurry face, low resolution, chaotic composition, blown facial highlights, cheap filters, random objects, duplicated person, unnatural neck or shoulders, or noisy artifacts.

查看 API 模型

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作