Datasets:
Tasks:
Image-Text-to-Video
Modalities:
Image
Formats:
imagefolder
Languages:
English
Size:
< 1K
ArXiv:
Add dataset card for RISE-Video
#1
by
nielsr
HF Staff
- opened
Hi! I'm Niels, part of the community science team at Hugging Face. I'm opening this PR to add a dataset card for RISE-Video.
The dataset card includes:
- Metadata for task category (
image-to-video), language, and the paper's ArXiv ID. - Links to the paper and the official GitHub repository.
- A description of the benchmark's 8 reasoning categories and 4 evaluation metrics.
- Usage instructions for video generation, frame extraction, and evaluation as described in the README.
- The BibTeX citation for the paper.
This will improve the visibility and usability of the benchmark on the Hugging Face Hub!