Papers › End-to-End Video Instance Segmentation with Transformers

End-to-End Video Instance Segmentation with Transformers

30 Nov 2020CVPR 2021 1arXiv:2011.14503archive 2025-07-28

Yuqing Wang, Zhaoliang Xu, Xinlong Wang, Chunhua Shen, Baoshan Cheng, Hao Shen, Huaxia Xia

Video instance segmentation (VIS) is the task that requires simultaneously classifying, segmenting and tracking object instances of interest in video. Recent methods typically develop sophisticated pipelines to tackle this task. Here, we propose a new video instance segmentation framework built upon Transformers, termed VisTR, which views the VIS task as a direct end-to-end parallel sequence decoding/prediction problem. Given a video clip consisting of multiple image frames as input, VisTR outputs the sequence of masks for each instance in the video in order directly. At the core is a new, effective instance sequence matching and segmentation strategy, which supervises and segments instances at the sequence level as a whole. VisTR frames the instance segmentation and tracking in the same perspective of similarity learning, thus considerably simplifying the overall pipeline and is significantly different from existing approaches. Without bells and whistles, VisTR achieves the highest speed among all existing VIS models, and achieves the best result among methods using single model on the YouTube-VIS dataset. For the first time, we demonstrate a much simpler and faster video instance segmentation framework built upon Transformers, achieving competitive accuracy. We hope that VisTR can motivate future research for more video understanding tasks.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Epiphqny/VisTR officialmentioned on GitHubpytorch report
spicyww/vistr_mindspore mentioned on GitHubmindspore report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Instance SegmentationSegmentationSemantic SegmentationVideo Instance SegmentationVideo Understanding

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-101) AP50 64.0 #25 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-101) AP75 45.0 #25 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-101) AR1 38.3 #25 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-101) AR10 44.9 #25 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-101) mask AP 40.1 #25 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-50) AP50 59.8 #30 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-50) AP75 36.9 #30 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-50) AR1 37.2 #30 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-50) AR10 42.4 #30 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation VisTR(ResNet-50) mask AP 36.2 #30 of 44 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: VisTR

VisTR

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections