MTEB logo MTEB
classification video

Kinetics700V

Cite this task

@article{smaira2020short,
  author = {Smaira, Lucas and Carreira, Joao and Noland, Eric and Clancy, Ellen and Wu, Amy and Zisserman, Andrew},
  journal = {arXiv preprint arXiv:2010.10864},
  title = {A Short Note on the Kinetics-700-2020 Human Action Dataset},
  year = {2020},
}

Kinetics-700-2020 is a large-scale action recognition dataset containing 700 human action classes from YouTube videos. Each clip is approximately 10 seconds long. This variant uses video only. Sampled the official test split, capped at ~16 per class across 700 classes (~11,190 examples).

Domains: ActivityWeb
Languages: English
Task type
VideoClassification
Main metric
accuracy
Source dataset
mteb/kinetics-700-2020
License
cc-by-4.0
Dates
2020-10-21 → 2020-10-21
Annotations
human-annotated
Sample creation
found
Models scored …
Dataset preview mteb/kinetics-700-2020 Open on HuggingFace →

Model scores

Loading…