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

清晨九宫格搞怪睡姿抓拍

Morning Nine-Grid Funny Sleeping Poses

一张9:16竖幅3×3清晨卧室抓拍拼图，以同一位明确成年女性九种搞怪、困倦的睡醒瞬间为主题。每格为极近距离随拍，保留自然凌乱的头发、枕头与被褥；所谓意外走光仅表现为睡衣或毯边轻微移位，人物始终完整穿着、非性化且不裸露。

清晨九宫格搞怪睡姿抓拍 — AI image creation
清晨九宫格搞怪睡姿抓拍

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9:16竖幅3×3拼图，呈现同一位明确成年女性在清晨卧室里九个搞怪、随意的睡觉或刚醒瞬间。九格均使用已配置ref-10a-v1保持同一面部身份。每格为极近距离特写，像早晨随手拍下的不完美私人照片；在人物和卧室保持连贯的前提下变化困倦搞笑瞬间。例如：脸颊被枕头压扁、头发翘起、被子盖住头只露出脸、困倦眨眼、枕头半遮住脸，或脸上压出枕头褶痕。不要出现手机UI，也不要声称拼图是真实相册截图。 女性每格都穿着遮盖完整的得体睡衣，整体非性化。所谓意外走光仅可表现为睡衣领口、袖口或毯边略微移位，露出仍有衣物遮挡的肩部或吊带；不得裸露、露出内衣或暴露私密部位。床单、枕头和柔和晨光卧室保持一致。采用自然清晨窗光、柔和阴影、真实皮肤与发丝纹理、随手拍构图、微妙颗粒和可信的抓拍瑕疵。九格表情与姿势清楚不同，氛围亲切好笑而非痛苦。不要增加其他人物、性化姿势、身体局部聚焦、裸露、乱码、标签、水印、边框或UI。

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 9:16 vertical 3-by-3 collage of nine funny, candid morning-bedroom sleeping and just-woken moments featuring the same clearly adult woman. Keep her facial identity consistent with configured ref-10a-v1 in every panel. Each square panel is an extreme close-up, like an imperfect personal snapshot taken in the morning; vary the comic sleepy moment while keeping one coherent person and bedroom. Examples: cheek squished into a pillow, hair sticking up, blanket pulled over the head with only the face peeking out, a groggy blink, a pillow partly covering the face, or an awkward pillow crease on the cheek. Do not show a phone UI or claim the collage is an authentic camera-roll screenshot. The woman wears modest, fully covering pajamas in every frame and remains nonsexual. Interpret any accidental exposure only as a pajama collar, sleeve or blanket edge shifting slightly to reveal a covered shoulder or camisole strap; no nudity, underwear reveal or exposed intimate areas. Keep the bed linens, pillows and softly lit morning bedroom as the consistent setting. Use natural early-morning window light, gentle shadows, realistic skin and hair texture, casual handheld framing, subtle grain and believable snapshot imperfections. Make the nine expressions/poses clearly distinct, funny and affectionate rather than distressing. No additional people, sexualized poses, body-focused framing, nudity, garbled text, labels, watermark, borders or UI.

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