metadata
license: cc-by-nc-4.0
task_categories:
- text-to-video
language:
- en
size_categories:
- 100K<n<1M
This is the dataset proposed in our paper "VidPro: A million-scale Prompt Dataset for Text-to-Video Diffusion Models"
It is the first and largest dataset for prompts used by text-to-video generation.
It is under construction.