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

Candlelit Tang-era shuanglu gathering

Five adult noblewomen share a lively late-night shuanglu game in a richly detailed Tang-inspired interior.

Candlelit Tang-era shuanglu gathering — AI image creation
Candlelit Tang-era shuanglu gathering

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。16:9 电影感室内中远景，五位成年唐代贵女围坐在低矮漆木双陆棋桌旁，参加深夜雅集。中央月白衣着的成年女性面部遵循 ref-10a-v1，神情平静带笑，双手靠近棋子；另外四位成年同伴均为彼此不同的虚构人物，身着有史感的唐代贵族服饰，观看、建议、打趣，其中一人调皮伸手拿棋子，引得众人发笑。服装含齐胸襦裙、宽袖、金步摇、玉簪与克制的花钿细节；五张脸彼此不同，不复制参考身份。无人直视镜头。场景为烛光映照的贵族内室，包含雕花木窗、漆艺屏风、锦缎帘幕、青铜烛台、香炉、茶杯和点心。以温暖琥珀色烛光映照偏冷环境暗部。保持指定配色：暖象牙白 #E7D8C4、蓝灰 #B8C0C2、漆木棕 #8D5D43、深影 #2E3238、烛光 #F0B66A、冷环境色 #6C7785。电影感柔和反差，火焰周围轻微光晕，细腻克制的颗粒与可信低照度柔和感。不要现代物件、电灯、奇幻服装、直视镜头、僵硬姿势、多余手指、融合的手、重复脸、水印、说明文字、字幕或漂浮文字。

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. 16:9 cinematic medium-wide interior, a late-night gathering of five adult Tang noblewomen around a low lacquered shuanglu board. The central woman, whose face follows ref-10a-v1, wears moon-white robes and smiles calmly, hands near the game pieces. Four distinct fictional adult companions in historically inspired Tang aristocratic attire watch, advise, tease, and share a laugh as one playfully reaches for a piece. Give them elegant high-waisted ruqun skirts, broad sleeves, gold buyao ornaments, jade hairpins, and restrained floral forehead details; no duplicate faces. No direct camera gaze. Set the scene in a candlelit aristocratic room with carved wooden windows, lacquered screens, brocade curtains, bronze candle stands, an incense burner, tea cups, and pastries. Use warm amber candlelight against cool ambient shadows. Maintain the supplied palette: warm ivory #E7D8C4, blue-gray #B8C0C2, lacquer brown #8D5D43, deep shadow #2E3238, candle glow #F0B66A, and cool ambient #6C7785. Soft cinematic contrast, gentle flame halation, subtle fine-grain texture, believable low-light softness. No modern objects, electric lights, fantasy costumes, direct gaze, stiff poses, extra fingers, fused hands, duplicate faces, watermark, captions, subtitles, or floating text.

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