# 1st Place Solution for YouTubeVOS Challenge 2021:Video Instance Segmentation

arXiv: https://arxiv.org/abs/2106.06649 · HTML: https://mlsanity.com/paper/2106.06649/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 25 references that are in the map.
- First posted on arXiv 12 Jun 2021.

## Builds on (most cited first, 15 of 15)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Deformable Convolutional Networks](https://mlsanity.com/paper/1703.06211/) (2017)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [ResNeSt: Split-Attention Networks](https://mlsanity.com/paper/2004.08955/) (2020)
- [End-to-End Video Instance Segmentation with Transformers](https://mlsanity.com/paper/2011.14503/) (2020)
- [Video Instance Segmentation](https://mlsanity.com/paper/1905.04804/) (2019)
- [Mask Scoring R-CNN](https://mlsanity.com/paper/1903.00241/) (2019)
- [Quasi-Dense Similarity Learning for Multiple Object Tracking](https://mlsanity.com/paper/2006.06664/) (2020)
- [Classifying, Segmenting, and Tracking Object Instances in Video with Mask Propagation](https://mlsanity.com/paper/1912.04573/) (2019)
- [Crossover Learning for Fast Online Video Instance Segmentation](https://mlsanity.com/paper/2104.05970/) (2021)
- [CompFeat: Comprehensive Feature Aggregation for Video Instance Segmentation](https://mlsanity.com/paper/2012.03400/) (2020)
- [1st Place Solutions for OpenImage2019 -- Object Detection and Instance Segmentation](https://mlsanity.com/paper/2003.07557/) (2020)

## Built on by (most cited first, 6 of 6)

- [SeqFormer: Sequential Transformer for Video Instance Segmentation](https://mlsanity.com/paper/2112.08275/) (2021)
- [In Defense of Online Models for Video Instance Segmentation](https://mlsanity.com/paper/2207.10661/) (2022)
- [Temporally Efficient Vision Transformer for Video Instance Segmentation](https://mlsanity.com/paper/2204.08412/) (2022)
- [CTVIS: Consistent Training for Online Video Instance Segmentation](https://mlsanity.com/paper/2307.12616/) (2023)
- [Tracking Instances as Queries](https://mlsanity.com/paper/2106.11963/) (2021)
- [VVitCutLER: Towards Unsupervised Object Detection and Segmentation in Videos](https://mlsanity.com/paper/2605.17584/) (2026)

## Abstract (arXiv)

Video Instance Segmentation (VIS) is a multi-task problem performing detection, segmentation, and tracking simultaneously. Extended from image set applications, video data additionally induces the temporal information, which, if handled appropriately, is very useful to identify and predict object motions. In this work, we design a unified model to mutually learn these tasks. Specifically, we propose two modules, named Temporally Correlated Instance Segmentation (TCIS) and Bidirectional Tracking (BiTrack), to take the benefit of the temporal correlation between the object's instance masks across adjacent frames. On the other hand, video data is often redundant due to the frame's overlap. Our analysis shows that this problem is particularly severe for the YoutubeVOS-VIS2021 data. Therefore, we propose a Multi-Source Data (MSD) training mechanism to compensate for the data deficiency. By combining these techniques with a bag of tricks, the network performance is significantly boosted compared to the baseline, and outperforms other methods by a considerable margin on the YoutubeVOS-VIS 2019 and 2021 datasets.