DAVIS and YouTube-VOS datasets

The DAVIS dataset comprises 60, 30 and 30 video sequences for training, validation and testing, respectively. The YouTube-VOS dataset is a larger dataset with 3471 videos for training and 507 for validation.

Data and Resources

Cite this as

Ali Athar, Jonathon Luiten, Alexander Hermans, Deva Ramanan, Bastian Leibe (2024). Dataset: DAVIS and YouTube-VOS datasets. https://doi.org/10.57702/8bi1egfz

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.1109/CVPR52688.2022.00303
Author Ali Athar
More Authors
Jonathon Luiten
Alexander Hermans
Deva Ramanan
Bastian Leibe
Homepage https://github.com/Ali2500/