Dataset · LightwheelAI
Egocentric video: 50-hour sample
LightwheelAI/EgoDemo
A 50-hour egocentric video sample for video model classification and evaluation.
What's inside
EgoDemo is a 50-hour sample from EgoSuite-Open100K. It includes all annotated subsets and two raw video variants. The English data is intended for video-classification; access is described via the EgoStandard and EgoPro collections and the project page.
What you can do with it
- Train a video classifier on egocentric clips
- Evaluate model quality on EgoSuite subsets
- Tasks
- Video classification
- Languages
- English
- Size
- 1K–10K rows
- Downloads this month
- 25.9K
- Likes
- 389
- Access
- on request on Hugging Face
How to load it
from datasets import load_dataset
ds = load_dataset("LightwheelAI/EgoDemo")