huggingface / huggingface/datasets
How to convert videos to images?
Open
enhancement
- Dominant language
- Python
- Stars
- 22k
- Forks
- 3.4k
- Avg merge
- 5d 7h
- Merged PRs (30d)
- 17
Description
### Feature request
Does someone know how to return the images from videos?
### Motivation
I am trying to use openpi(https://github.com/Physical-Intelligence/openpi) to finetune my Lerobot dataset(V2.0 and V2.1). I find that although the codedaset is v2.0, they are different. It seems like Lerobot V2.0 has two version, one is data include images infos and another one is separate to data and videos.
Does someone know how to return the images from videos?
Contributor guide
Assessment
This issue has not been assessed yet.