Papers › DVIS: Decoupled Video Instance Segmentation Framework
DVIS: Decoupled Video Instance Segmentation Framework
6 Jun 2023ICCV 2023 1arXiv:2306.03413archive 2025-07-28
Tao Zhang, Xingye Tian, Yu Wu, Shunping Ji, Xuebo Wang, Yuan Zhang, Pengfei Wan
Video instance segmentation (VIS) is a critical task with diverse applications, including autonomous driving and video editing. Existing methods often underperform on complex and long videos in real world, primarily due to two factors. Firstly, offline methods are limited by the tightly-coupled modeling paradigm, which treats all frames equally and disregards the interdependencies between adjacent frames. Consequently, this leads to the introduction of excessive noise during long-term temporal alignment. Secondly, online methods suffer from inadequate utilization of temporal information. To tackle these challenges, we propose a decoupling strategy for VIS by dividing it into three independent sub-tasks: segmentation, tracking, and refinement. The efficacy of the decoupling strategy relies on two crucial elements: 1) attaining precise long-term alignment outcomes via frame-by-frame association during tracking, and 2) the effective utilization of temporal information predicated on the aforementioned accurate alignment outcomes during refinement. We introduce a novel referring tracker and temporal refiner to construct the Decoupled VIS framework (DVIS). DVIS achieves new SOTA performance in both VIS and VPS, surpassing the current SOTA methods by 7.3 AP and 9.6 VPQ on the OVIS and VIPSeg datasets, which are the most challenging and realistic benchmarks. Moreover, thanks to the decoupling strategy, the referring tracker and temporal refiner are super light-weight (only 1.69% of the segmenter FLOPs), allowing for efficient training and inference on a single GPU with 11G memory. The code is available at https://github.com/zhang-tao-whu/DVIS.
In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.
Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
For agents, on Syntology's MCP service (how to connect):
get_citation_path(paper_1="2306.03413", paper_2="…") with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2306.03413")
Code
zhang-tao-whu/DVIS
officialmentioned in papermentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries (arXiv:2404.00086; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. 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.
A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
1 archive task tag without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Offline) |
AP50 |
75.9 |
#5 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Offline) |
AP75 |
53.0 |
#5 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Offline) |
AR1 |
19.4 |
#5 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Offline) |
AR10 |
55.3 |
#5 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Offline) |
mask AP |
49.9 |
#5 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Online) |
AP50 |
71.9 |
#8 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Online) |
AP75 |
49.2 |
#8 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Online) |
AR1 |
19.4 |
#8 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Online) |
AR10 |
52.5 |
#8 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS(Swin-L, Online) |
mask AP |
47.1 |
#8 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS(Swin-L) |
AP50 |
83.0 |
#8 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS(Swin-L) |
AP75 |
68.4 |
#8 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS(Swin-L) |
AR1 |
47.7 |
#8 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS(Swin-L) |
AR10 |
65.7 |
#8 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS(Swin-L) |
mask AP |
60.1 |
#8 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS |
AP50 |
88.0 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS |
AP75 |
72.7 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS |
AR1 |
56.5 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS |
AR10 |
70.3 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS |
mask AP |
64.9 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS(Swin-L) |
AP50_L |
69.0 |
#4 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS(Swin-L) |
AP75_L |
48.8 |
#4 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS(Swin-L) |
AR10_L |
51.8 |
#4 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS(Swin-L) |
AR1_L |
37.2 |
#4 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS(Swin-L) |
mAP_L |
45.9 |
#4 of 7 |
Archive leaderboard |
report |
| Video Panoptic Segmentation |
VIPSeg |
DVIS(Swin-L) |
STQ |
55.3 |
#4 of 12 |
Archive leaderboard |
report |
| Video Panoptic Segmentation |
VIPSeg |
DVIS(Swin-L) |
VPQ |
57.6 |
#4 of 12 |
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.
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