该提示词用于分析剧本并提取所有出场人物和核心场景,为每个角色生成包含姓名、年龄、外貌、服装、气质与绘画提示词的角色设定,为场景生成概念图提示词,并列出需保持一致的重要道具,适合短剧前期美术资产准备。

中文版提示词

请分析以上剧本,提取所有出场人物和核心场景,并生成 AI 绘画可用的中文提示词。
人物部分:每个角色包含姓名、年龄、身份、人物关系、外貌特征、发型发色、服装风格、核心气质、常见表情、关键动作,以及一段人物绘画提示词。人物绘画提示词要求:全身远景,正面站立,纯色背景,真人写实电影质感,角色设定表,服装面料和面部特征清晰,不要文字、不要水印、不要多余人物。
场景部分:每个场景包含场景名称、出现剧情、时间、空间结构、光影氛围、材质细节、镜头视角,以及一段场景绘画提示词。场景绘画提示词要求:场景概念图,真人短剧质感,真实物理光影,空间层次清晰,电影级构图,不要人物、不要文字、不要水印。
道具部分:请列出剧中需要保持一致的重要道具(比如戒指、合同、手机、录音笔、文件袋、车钥匙等),并给出外观描述。

英文版提示词

Analyze the script above, extract all characters and core scenes, and generate Chinese prompts usable for AI drawing.
Characters: each character should include name, age, identity, relationships, appearance, hairstyle and color, clothing style, core temperament, common expressions, key actions, and a character-drawing prompt. Character-prompt requirements: full-body long shot, front-facing standing pose, solid-color background, photorealistic cinematic quality, character-sheet style, clear clothing fabric and facial features, no text, no watermark, no extra characters.
Scenes: each scene should include scene name, plot context, time, spatial structure, lighting atmosphere, material details, camera angle, and a scene-drawing prompt. Scene-prompt requirements: scene concept art, realistic short-drama quality, realistic physical lighting, clear spatial layering, cinematic composition, no people, no text, no watermark.
Props: list the important props that must stay consistent throughout (e.g., ring, contract, phone, voice recorder, file folder, car keys, etc.) and describe their appearance.

🛠️ **适用 AI 工具**:Midjourney、Stable Diffusion、DALL·E、即梦、豆包、Flux