News

  • The 2nd Large-scale Video Object Segmentation Challenge has finished. Congrats Zhishan Zhou’s team for winning the VOS track and Jonathon Luiten’s team for winning the VIS track! Leaderboard has been updated.
  • We will hold “The 2nd Large-scale Video Object Segmentation Challenge” in conjunction with ICCV 2019 in Seoul, Korea. Stay tuned!

What is YouTube-VOS

YouTube-VOS is the first large-scale benchmark that supports multiple video object segmentation tasks.

  • Semi-supervised Video Object Segmentation
  • Video Instance Segmentation

It also has the following features.

  • 4000+ high-resolution YouTube videos
  • 90+ semantic categories
  • 7800+ unique objects
  • 190k+ high-quality manual annotations
  • 340+ minutes duration

Research paper

Please cite the following papers if you find our dataset is useful.

Semi-supervised video object segmentation

Video instance segmentation

Dataset examples