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

暖光卧室里的三格自拍拼贴

Three-Panel Selfie Collage in a Warmly Lit Bedroom

同一位成年女性的三格竖版自拍拼贴：在柔和阳光卧室中分别做剪刀手、双手托腮和OK手势，周围点缀粉彩手绘涂鸦，呈现自然、轻松的手机照片质感。

暖光卧室里的三格自拍拼贴 — AI image creation
暖光卧室里的三格自拍拼贴

提示词 / Prompt

中文提示词

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。 创作一张写实的竖版三格自拍拼贴，三个面板中始终是同一位明确成年的女性。她的面部与发型仅依据已附加参考图。她留着飘逸长发，穿着休闲纯白T恤，保留自然皮肤纹理和轻松愉快的表情。第一格比出剪刀手，第二格用双手轻托脸颊，第三格做出OK手势。三格的取景与人物比例保持协调，让画面读起来像同一人连续拍下的一组俏皮自拍。 场景为柔和日照的卧室，采用soft warm afternoon window light。在人物周围点缀小巧的粉彩手绘爱心、星星、笑脸气泡和可爱轮廓线，不遮挡脸部，使用pastel pink, lavender, and pale yellow。加入subtle fine film grain、轻微褪色的色彩和真实手机抓拍质感。整体温暖、梦幻、青春，但人物必须呈现为成年人。 反向提示词：动漫、卡通脸、塑料感或过度磨皮的皮肤、虚假美颜滤镜、扭曲的手、多余手指、畸形结构、浓妆、刺眼阴影、低分辨率、乱码、标志、水印、签名、其他人物、三格之间面部不一致。

English prompt

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 realistic vertical three-panel selfie collage showing the same clearly adult woman in every panel. Preserve her face and hairstyle solely from the attached reference. She has long flowing hair and wears a casual plain white T-shirt, with natural skin texture and relaxed, cheerful expressions. In the first panel she makes a peace sign; in the second she gently rests both hands against her cheeks; in the third she makes an OK sign. Keep the framing and scale coherent across all three panels so the collage reads as one person taking a playful sequence of selfies. Set the scene in a softly sunlit bedroom with soft warm afternoon window light. Add small pastel hand-drawn hearts, stars, smiley bubbles and cute outline doodles around—not over—the face, using pastel pink, lavender, and pale yellow. Use subtle fine film grain, gently washed colors and the candid look of a real phone photo. Keep the image warm, dreamy and youthful without making the subject a minor. Negative prompt: anime, cartoon face, plastic or over-smoothed skin, beauty-filter look, distorted hands, extra fingers, malformed anatomy, heavy makeup, harsh shadows, low resolution, garbled text, logo, watermark, signature, extra people, inconsistent face between panels.

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