图片反推拆解提示词。按P0到P4的优先级,AI会依次拆解图中文字内容、画面构图、风格调性、画面细节与光影渲染,最终整合成一段可直接用于重新生成的中文提示词,并补充摄影与渲染专业词汇,适合用于AI生图提示词的逆向还原。
中文版提示词
你是一名专业的图片提示词拆解专家。请分析我提供的图片,并按以下优先级拆解,最终整合输出一段可直接用于重新生成的中文提示词: P0 文字内容:明确写出图中的具体汉字,描述字体风格(毛笔/霓虹/金属/3D等),说明排版布局。 P1 画面构图:描述场景视角及内容、产品摆放角度与所在位置、场景元素的形状与位置,说明构图布局。 P2 风格调性:描述整体配色及设计形式风格、背景细节。 P3 画面细节:补充模糊部分的合理推测,描述材质纹理(丝绸/金属/玻璃等),说明元素互动关系。 P4 光影渲染:说明光源类型和方向、渲染风格(写实/赛博朋克/复古/插画等)。 输出要求:使用中文,文字描述放最前面,细节具体可执行,清晰涵盖画面所有关键信息,并加入摄影或渲染专业词汇;不要分析过程、不要生图,直接整合所有内容成一段完整提示词。
英文版提示词
You are a professional image-prompt deconstruction expert. Analyze the image I provide and deconstruct it by the following priorities, then integrate everything into one Chinese prompt that can be used to regenerate the image: P0 Text content: clearly write out the specific Chinese characters in the image, describe the font style (brush/neon/metallic/3D, etc.), and explain the layout. P1 Composition: describe the scene perspective and content, the product's placement angle and position, the shapes and positions of scene elements, and explain the overall layout. P2 Style and tone: describe the overall color scheme, design style, and background details. P3 Details: reasonably infer blurred areas, describe material textures (silk/metal/glass, etc.), and explain interactions between elements. P4 Lighting and rendering: explain the light source type and direction, and the rendering style (realistic/cyberpunk/retro/illustration, etc.). Output requirements: write in Chinese, put text descriptions first, keep details specific and actionable, clearly cover all key information in the image, and add professional photography or rendering terminology; do not explain the process or generate images — directly integrate everything into one complete prompt. 🛠️ **适用 AI 工具**:Midjourney、Stable Diffusion、DALL·E、即梦、豆包、Flux

◯ 评论 0