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

暖金室内光线下的三联内省肖像

Triptych of Introspective Portraits in Warm Golden Indoor Light

同一位绑定固定身份参考的成年人物以9:16竖版三联画呈现三个安静室内瞬间，穿温暖中性色针织开衫与长裤，在书架、沙发和复古木椅间由金色日光勾勒出内省的电影感氛围。

暖金室内光线下的三联内省肖像 — AI image creation
暖金室内光线下的三联内省肖像

提示词 / Prompt

中文提示词

创作一张9:16竖版电影感写实照片拼贴，恰好由三个纵向叠放的场景组成，三格均为同一位成年人物，使用固定身份参考ref-10a-v1。面部身份、发型与服装在各格保持一致。人物穿柔软优雅的休闲叠搭：温暖中性色针织开衫、简洁内搭与舒适长裤，不复制不可见来源照片中的服装。第一格：人物在木书架旁的侧脸特写，头部轻轻低垂，神情平静内省；温暖金色侧后方日光为发丝勾出柔和轮廓光，并在面部留下浅淡阴影。第二格：较宽景别中，人物自然坐在沙发与素墙附近的地面上，双膝舒适地靠近，氛围安静孤独，日落光形成深琥珀色阴影。第三格：中景呈现人物自然斜靠在复古躺椅或木椅上，微微向上或望向远处，神情沉思，柔和背光勾勒头发与肩膀。三段画面分隔清楚，同时构成统一亲密的室内故事。使用温暖室内黄金时段光线、浅景深、真实皮肤纹理、自然姿态、柔和发丝细节、胶片感调色和精致编辑肖像风格。不得出现文字、标志、水印、其他人物、重复畸形肢体或特定相机/分辨率承诺。

English prompt

Create a realistic 9:16 vertical cinematic photo collage with exactly three vertically stacked scenes featuring the same adult person, using the fixed identity reference ref-10a-v1. Keep facial identity, hairstyle, and outfit consistent in all three panels. Dress the person in soft, elegant casual layers: a warm-neutral knit cardigan over a simple top with comfortable matching trousers, different from any unseen source clothing. Panel 1: a close side-profile portrait beside a wooden bookshelf, head gently lowered and expression calm and introspective; warm golden light from the side and back gives the hair a soft rim glow and leaves gentle facial shadows. Panel 2: a wider view of the person sitting naturally on the floor near a sofa and plain wall, knees drawn comfortably close, with a quiet, solitary mood and deep amber sunset shadows. Panel 3: a medium view reclining naturally in a vintage lounge chair or wooden chair, looking slightly upward or into the distance with a thoughtful expression, softly backlit around the hair and shoulders. Separate the three moments cleanly while keeping a unified intimate indoor story. Use warm indoor golden-hour light, shallow depth of field, realistic skin texture, natural posture, soft hair detail, film-like color grading, and refined editorial portrait styling. No text, logos, watermark, extra people, duplicated anatomy, or specific camera/resolution claims.

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