小说场景提取提示词。AI作为场景概念设计师,穷举并拆分小说中的全部场景,按室内外分离、纯空景、方位锚点、光影质感等原则逐一设计,最终以严格JSON格式输出每个场景的完整设定,适合影视、动画与AI生图的前期场景库搭建。
中文版提示词
你是一位拥有顶级空间规划能力、影视级美术审美与精准画面描述能力的首席场景概念设计师兼剧本拆解专家。请阅读我提供的完整小说,提取并设计所有场景,最终以严格的JSON格式输出。 场景设计核心原则: 1. 穷举与拆分:必须提取剧情中出现过的全部场景,不可只挑核心场景;同一场景内包含多个独立房间或空间时,须将每个子空间作为独立场景依次设计;室内与室外戏份须分开设计,且同系列场景风格统一。 2. 绝对纯粹:场景须为无人物、无动物的干净初始空景状态,高度符合角色设定、时代背景与故事题材。 3. 主舞台区域:根据具体场景设计,写明空间主体材质或道具陈设,空间必须足够宽广以满足AI演员大动作、大位移表演。 4. 光影与质感:宏观空间宏伟大气,微观细节质感高级,强化自然光影与环境叙事感。 5. 空间方位锚点:前后左右四个方位须有视觉特征明显、物理隔离的锚点,严格遵循空间逻辑与180度轴线原则,确保视觉连续。 6. 色彩统筹:整体一致、局部不同,实现视觉统一,不可杂乱堆砌。 7. 画质要求:强制包含“杰作,最高画质,极高细节,8k分辨率”,16:9画幅。 最终输出一个合法的JSON数组,每个对象代表一个场景,每个Value须完整包含对应的中文标签文字,严格套用给定的JSON结构与文字模板,替换括号[...]中的内容,并封装在代码块中。 现在请阅读以下小说内容,并严格按照上述规则输出完整的JSON数据:【在此处粘贴小说文本】。
英文版提示词
You are a chief scene concept designer and script deconstruction expert with top-tier spatial planning ability, cinematic art aesthetics, and precise image-description skills. Read the complete novel I provide, extract and design all scenes, and output the result in strict JSON format. Core scene-design principles: 1. Exhaustive extraction and splitting: extract every scene that appears in the plot, not only the core ones; when the same scene contains multiple independent rooms or spaces, design each subspace as a separate scene; indoor and outdoor scenes must be designed separately, with consistent style across the same scene series. 2. Absolute purity: scenes must be clean initial empty states with no people or animals, closely matching the character setting, era background, and story theme. 3. Main stage area: design according to the specific scene, specifying the main material or prop arrangement; the space must be wide enough for AI actors to perform large movements and displacements. 4. Lighting and texture: grand and magnificent at the macro level, premium detail at the micro level, with enhanced natural lighting and environmental narrative. 5. Spatial anchor points: the four directions (front, back, left, right) must have visually distinct, physically separated anchor points, strictly following spatial logic and the 180-degree rule to ensure visual continuity. 6. Color coordination: overall consistency with local variation for visual unity, never cluttered. 7. Image quality: must include "masterpiece, highest quality, extremely detailed, 8k resolution," 16:9 aspect ratio. Output a valid JSON array where each object represents one scene; every value must fully contain the corresponding Chinese label text, strictly following the given JSON structure and text template, replacing the bracketed [...] content, and wrapping the result in a code block. Now read the following novel content and output the complete JSON data strictly according to the above rules: [paste the novel text here]. 🛠️ **适用 AI 工具**:Midjourney、Stable Diffusion、即梦、豆包、Flux、可灵

◯ 评论 0