Three-Asset Composite — Person From Image 1, Scene From Image 2, Music From Audio 1

Seedance 2.0CreativeStructured PromptImage to VideoMusic Video15s
The person from [Image 1] dances in the setting of [Image 2] to the track in [Audio 1] — a choreographed prompt referencing three modalities at once, a minimal expression of multi-asset workflows.
PROMPT · Seedance 2.0 提示词
Create with AI
A woman from [Image 1] is performing a dope hip-hop dance in the location from [Image 2], set to the music of [Audio 1]. Explode with creativity and express a 15-second music video never before seen in this world. Whether it's a multi-cut or a long single take is known only to God. An artistic 15-second MV that excites viewers, influenced by Fauvist and Surrealist artists.
✍️ Editor’s Notes

这条不是纯文字prompt,而是人物图、场景图、音频各给一份的三件套输入,文字部分只负责把三者粘合成叙事逻辑,「from [Image 1]」「in the location from [Image 2]」这种写法是Seedance多模态参考能力的标准调用方式。后半句「whether it's a multi-cut or a long single take is known only to God」看着像玩笑,其实是故意不锁运镜方式,把镜头调度这个变量交给模型自由发挥,减少过度控制反而更容易出彩。

Create with AI
Related Works
×