PROMPT RADAR / VIDEO视频叙事
THINK YOU NEED A BIG BUDGET TO MAKE KIDS’ VIDEOS ON YOUTUBE??? THIS TEEN
THINK YOU NEED A BIG BUDGET TO MAKE KIDS’ VIDEOS ON YOUTUBE??? THIS TEEN BUILDS ONE IN UNDER 60 SECONDS. the entire workflow starts inside a chat window with a single image prompt. here’s the process: → open ChatGPT and
@He1s_Sammy 1.1k 热度 1.1 万 浏览
编辑先看
它为什么值得进入灵感库
它最有价值的部分，是把一个模糊想法推进到了可以被模型执行的程度。 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
THINK YOU NEED A BIG BUDGET TO MAKE KIDS’ VIDEOS ON YOUTUBE??? THIS TEEN BUILDS ONE IN UNDER 60 SECONDS. the entire workflow starts inside a chat window with a single image prompt. here’s the process: → open ChatGPT and paste the character + scene prompt → send it into Picsart Playground → generate bright, kid-friendly visuals in seconds → combine a few frames into a simple story → add a soft nursery track → turn it into a complete video without filming anything no camera. no studio. no animation team. just a simple AI-powered production workflow. once a channel finds a format that kids actually want to watch and replay, one person can keep producing and publishing without ever stepping in front of a camera. the exact prompt and full 60 second workflow are quoted below. bookmark this before you lose it.
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：THINK YOU NEED A BIG BUDGET TO MAKE KIDS’ VIDEOS ON YOUTUBE??? THIS TEEN BUILDS ONE IN UNDER 60 SECONDS. the entire workflow starts inside a chat window with a single image prompt. here’s the process: → open ChatGPT and
- 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