AI Chat AI Image Generator AI Video Text to Speech

The Runner-up Solution for YouTube-VIS Long Video Challenge 2022

11/18/2022

∙

by Junfeng Wu, et al.

∙

∙

This technical report describes our 2nd-place solution for the ECCV 2022 YouTube-VIS Long Video Challenge. We adopt the previously proposed online video instance segmentation method IDOL for this challenge. In addition, we use pseudo labels to further help contrastive learning, so as to obtain more temporally consistent instance embedding to improve tracking performance between frames. The proposed method obtains 40.2 AP on the YouTube-VIS 2022 long video dataset and was ranked second place in this challenge. We hope our simple and effective method could benefit further research.

Junfeng Wu
14 publications
Yi Jiang
39 publications
Qihao Liu
10 publications
Xiang Bai
128 publications
Song Bai
72 publications

research

∙ 06/20/2022

5th Place Solution for YouTube-VOS Challenge 2022: Video Object Segmentation

Video object segmentation (VOS) has made significant progress with the r...

0 Wangwang Yang, et al. ∙

research

∙ 06/22/2021

Tracking Instances as Queries

Recently, query based deep networks catch lots of attention owing to the...

0 Shusheng Yang, et al. ∙

research

∙ 12/13/2021

5th Place Solution for VSPW 2021 Challenge

In this article, we introduce the solution we used in the VSPW 2021 Chal...

0 Jiafan Zhuang, et al. ∙

research

∙ 07/21/2022

In Defense of Online Models for Video Instance Segmentation

In recent years, video instance segmentation (VIS) has been largely adva...

1 Junfeng Wu, et al. ∙

research

∙ 01/15/2019

Measuring Effectiveness of Video Advertisements

Advertisements are unavoidable in modern society. Times Square is notori...

4 James Hahn, et al. ∙

research

∙ 12/27/2022

1st Place Solution for YouTubeVOS Challenge 2022: Referring Video Object Segmentation

The task of referring video object segmentation aims to segment the obje...

0 Zhiwei Hu, et al. ∙