PROMPT RADAR / VIDEO视频叙事
Seedance 2.0 Prompt: Generate a seamless continuous one-take video using
Seedance 2.0 Prompt: Generate a seamless continuous one-take video using the reference images. No dialogue, hard cuts, fade transitions, or opacity changes at any moment. Use ultra slow motion with a relaxed summer atmos
@Naiknelofar788 445 热度 3.2k 浏览
编辑先看
它为什么值得进入灵感库
这条内容值得留下的,不是表面的风格词,而是它提供了一条可以继续实验的画面路径。 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
Seedance 2.0 Prompt: Generate a seamless continuous one-take video using the reference images. No dialogue, hard cuts, fade transitions, or opacity changes at any moment. Use ultra slow motion with a relaxed summer atmosphere and nostalgic ’90s visual aesthetic. Connect every scene naturally using only smooth camera movement transitions. Add matching transition sound effects and keep the cinematic ambient audio flowing continuously without interruption. No music. Maintain visual consistency, coherent motion, and realistic cinematic lighting throughout. 15 seconds, 16:9 aspect ratio, 720p resolution. The camera movement was the most important part. Instead of asking the model to switch between scenes, I described every transition as a physical camera action, such as pushing through an object, panning past a foreground element, or following the subject into the next environment.
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:Seedance 2.0 Prompt: Generate a seamless continuous one-take video using the reference images. No dialogue, hard cuts, fade transitions, or opacity changes at any moment. Use ultra slow motion with a relaxed summer atmos
- 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