Introduction

Our workshop has three challenges for different video segmentation tasks including semi-supervised video object segmentation, video instance segmentation and referring video object segmentation. The first two challenges will be very similar to previous challenges we have done with improved and augmented datasets (details). The third challenge aims to segment an object referred by a given language expression in a video and requires algorithms to understand video and language jointly.

Announcement

Dates

  • May 20th: The final competition results will be announced and top teams will be invited to give oral/poster presentations at our CVPR 2021 workshop.
  • May 5th - 14th: Release test data and open the submission of the test results.
  • Feb 15th: Codalab websites open for registration. Training and validation data are released.

Tasks

Submission

Organizers

Ning Xu Linjie Yang Yuchen Fan Yang Fu Weiyao Lin
Ning Xu
Adobe Research
Linjie Yang
ByteDance AI Lab
Yuchen Fan
UIUC
Yang Fu
UIUC
Weiyao Lin
SJTU, China
Jianchao Yang Honghui Shi Joon-Young Lee Seonguk Seo  
Jianchao Yang
ByteDance AI Lab
Humphrey Shi
UIUC
Joon-Young Lee
Adobe Research
Seonguk Seo
SNU, Korea
 

Contact

For dataset related questions, please feel free to contact ytbvos@gmail.com. For challenge related questions, you can also use codalab forums.

Sponsors

Adobe ByteDance UIUC
Adobe ByteDance UIUC