PROMPT RADAR / VIDEO视频叙事
She has never held a camera or drawn a single frame. She just cleared $2
She has never held a camera or drawn a single frame. She just cleared $20,000 in 30 days from AI-generated cartoons, published from an iPad by the pool. The video is the exact workflow, start to finish. She picks a kids
@0x_Tommy_ 410 热度 9.6k 浏览
编辑先看
它为什么值得进入灵感库
先别急着照抄整段 Prompt：真正有用的是主体、空间和镜头之间的约束关系。 把一段描述拆成起始状态、动作变化、镜头响应和结束画面，连续性比形容词更重要。
01 / 原始方法
动作与镜头 Prompt
She has never held a camera or drawn a single frame. She just cleared $20,000 in 30 days from AI-generated cartoons, published from an iPad by the pool. The video is the exact workflow, start to finish. She picks a kids channel already pulling millions of views, feeds the transcript into AI, and asks the model to build an original show around what already keeps children watching. Song structure, pacing, repeating visual patterns. That prompt goes into a video generator. Ten minutes later her iPad has a finished 4K cartoon with animated characters, voices, music, sound effects, ready to publish. The real edge is not making one AI cartoon. It is that one working format becomes dozens of episodes off the same chair by the pool. Save this before your feed fills up with people running the same loop from their kitchen table.
02 / 复刻路径
不要一次改完所有变量
用 Seedance 复刻时，建议先单独生成稳定首帧，再把动作拆成“开始、变化、收束”三段；每段只保留一个主要运动。
- 主体：She has never held a camera or drawn a single frame. She just cleared $20,000 in 30 days from AI-generated cartoons, published from an iPad by the pool. The video is the exact workflow, start to finish. She picks a kids
- 视觉方向：实时 AI 视频案例；查看原帖媒体与作者说明
- 镜头与构图：按原帖镜头流程执行；查看原帖媒体与工作流
- 交付约束：auto，保留原作者方法与归属；复刻时避免水印、低清、畸变、主体漂移和无关文字。
负向约束
保留原作者方法与归属；复刻时避免水印、低清、畸变、主体漂移和无关文字。
模型建议
Seedance / Veo / Kling。先用原始 Prompt 建立基准结果，再根据模型对自然语言、镜头运动或文字排版的能力做局部改写。
来源与编辑原则