Papers › DVIS++: Improved Decoupled Framework for Universal Video Segmentation
DVIS++: Improved Decoupled Framework for Universal Video Segmentation
20 Dec 2023arXiv:2312.13305archive 2025-07-28
Tao Zhang, Xingye Tian, Yikang Zhou, Shunping Ji, Xuebo Wang, Xin Tao, Yuan Zhang, Pengfei Wan, Zhongyuan Wang, Yu Wu
We present the Decoupled VIdeo Segmentation (DVIS) framework, a novel approach for the challenging task of universal video segmentation, including video instance segmentation (VIS), video semantic segmentation (VSS), and video panoptic segmentation (VPS). Unlike previous methods that model video segmentation in an end-to-end manner, our approach decouples video segmentation into three cascaded sub-tasks: segmentation, tracking, and refinement. This decoupling design allows for simpler and more effective modeling of the spatio-temporal representations of objects, especially in complex scenes and long videos. Accordingly, we introduce two novel components: the referring tracker and the temporal refiner. These components track objects frame by frame and model spatio-temporal representations based on pre-aligned features. To improve the tracking capability of DVIS, we propose a denoising training strategy and introduce contrastive learning, resulting in a more robust framework named DVIS++. Furthermore, we evaluate DVIS++ in various settings, including open vocabulary and using a frozen pre-trained backbone. By integrating CLIP with DVIS++, we present OV-DVIS++, the first open-vocabulary universal video segmentation framework. We conduct extensive experiments on six mainstream benchmarks, including the VIS, VSS, and VPS datasets. Using a unified architecture, DVIS++ significantly outperforms state-of-the-art specialized methods on these benchmarks in both close- and open-vocabulary settings. Code: https://github.com/zhang-tao-whu/DVIS_Plus.
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="2312.13305", paper_2="…") with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2312.13305")
Code
zhang-tao-whu/DVIS_Plus
officialmentioned in papermentioned on GitHubpytorchSyntology: 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
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L,Offline) |
AP50 |
78.9 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L,Offline) |
AP75 |
58.5 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L,Offline) |
mask AP |
53.4 |
#3 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
AP50 |
72.5 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
AP75 |
55.0 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
APho |
27.1 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
APmo |
56.6 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
APso |
69.9 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
AR1 |
20.8 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
AR10 |
54.6 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(VIT-L, Online) |
mask AP |
49.6 |
#6 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Offline) |
AP50 |
68.9 |
#19 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Offline) |
AP75 |
40.9 |
#19 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Offline) |
AR1 |
16.8 |
#19 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Offline) |
AR10 |
47.3 |
#19 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Offline) |
mask AP |
41.2 |
#19 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Online) |
AP50 |
62.8 |
#22 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Online) |
AP75 |
37.3 |
#22 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Online) |
AR1 |
15.8 |
#22 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Online) |
AR10 |
42.9 |
#22 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
OVIS validation |
DVIS++(R50, Online) |
mask AP |
37.2 |
#22 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Offline) |
AP50 |
86.7 |
#3 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Offline) |
AP75 |
71.5 |
#3 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Offline) |
AR1 |
48.8 |
#3 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Offline) |
AR10 |
69.5 |
#3 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Offline) |
mask AP |
63.9 |
#3 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Online) |
AP50 |
82.7 |
#4 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Online) |
AP75 |
70.2 |
#4 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Online) |
AR1 |
49.5 |
#4 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Online) |
AR10 |
68.0 |
#4 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS 2021 |
DVIS++(VIT-L, Online) |
mask AP |
62.3 |
#4 of 26 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS++(ViT-L, Online) |
AP50 |
88.8 |
#2 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS++(ViT-L, Online) |
AP75 |
75.3 |
#2 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS++(ViT-L, Online) |
AR1 |
57.9 |
#2 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS++(ViT-L, Online) |
AR10 |
73.7 |
#2 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
YouTube-VIS validation |
DVIS++(ViT-L, Online) |
mask AP |
67.7 |
#2 of 44 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS++(VIT-L) |
AP50_L |
75.7 |
#1 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS++(VIT-L) |
AP75_L |
52.8 |
#1 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS++(VIT-L) |
AR10_L |
55.8 |
#1 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS++(VIT-L) |
AR1_L |
40.6 |
#1 of 7 |
Archive leaderboard |
report |
| Video Instance Segmentation |
Youtube-VIS 2022 Validation |
DVIS++(VIT-L) |
mAP_L |
50.9 |
#1 of 7 |
Archive leaderboard |
report |
| Video Panoptic Segmentation |
VIPSeg |
DVIS++(VIT-L) |
STQ |
56.0 |
#2 of 12 |
Archive leaderboard |
report |
| Video Panoptic Segmentation |
VIPSeg |
DVIS++(VIT-L) |
VPQ |
58.0 |
#2 of 12 |
Archive leaderboard |
report |
| Video Semantic Segmentation |
VSPW |
DVIS++(VIT-L) |
mIoU |
63.8 |
#1 of 5 |
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
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