Video Instance Segmentation on YouTube-VIS 2021

Metric: AR1 (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	AR1▼	Extra Data	Paper	Date↕	Code
1	CAVIS(VIT-L, Offline)	49.7	Yes	Context-Aware Video Instance Segmentation	2024-07-03	Code
2	DVIS-DAQ(VIT-L, Offline)	49.6	Yes	DVIS-DAQ: Improving Video Segmentation via Dynam...	2024-03-29	Code
3	DVIS++(VIT-L, Online)	49.5	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
4	GenVIS (Swin-L)	49.1	Yes	A Generalized Framework for Video Instance Segme...	2022-11-16	Code
5	DVIS++(VIT-L, Offline)	48.8	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
6	GRAtt-VIS (Swin-L)	48.8	Yes	GRAtt-VIS: Gated Residual Attention for Auto Rec...	2023-05-26	Code
7	RefineVIS (Swin-L, online)	48.3	Yes	RefineVIS: Video Instance Segmentation with Temp...	2023-06-07	-
8	NOVIS (Swin-L)	47.9	Yes	NOVIS: A Case for End-to-End Near-Online Video I...	2023-08-29	-
9	DVIS(Swin-L)	47.7	Yes	DVIS: Decoupled Video Instance Segmentation Fram...	2023-06-06	Code
10	VITA (Swin-L)	47.7	Yes	VITA: Video Instance Segmentation via Object Tok...	2022-06-09	Code
11	TarViS (Swin-L)	47.6	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
12	Tube-Link(Swin-L)	47.5	No	Tube-Link: A Flexible Cross Tube Framework for U...	2023-03-22	Code
13	UniVS(Swin-L)	46.2	Yes	UniVS: Unified and Universal Video Segmentation ...	2024-02-28	Code
14	MinVIS (Swin-L)	45.9	No	MinVIS: A Minimal Video Instance Segmentation Fr...	2022-08-03	Code
15	MDQE(Swin-L)	45.4	No	MDQE: Mining Discriminative Query Embeddings to ...	2023-03-25	Code
16	IDOL (Swin-L)	45	No	In Defense of Online Models for Video Instance S...	2022-07-21	Code
17	BoxVIS(Swin-L & Box-sup)	44.8	No	BoxVIS: Video Instance Segmentation with Box Ann...	2023-03-26	Code
18	DeVIS (Swin-L)	43.8	No	DeVIS: Making Deformable Transformers Work for V...	2022-07-22	Code
19	InstanceFormer (Swin-L)	42.8	Yes	InstanceFormer: An Online Video Instance Segment...	2022-08-22	Code
20	TarViS (Swin-T)	42.2	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
21	GRAtt-VIS (ResNet-50)	41.8	Yes	GRAtt-VIS: Gated Residual Attention for Auto Rec...	2023-05-26	Code
22	NOVIS (ResNet-50)	41.3	Yes	NOVIS: A Case for End-to-End Near-Online Video I...	2023-08-29	-
23	TarViS (ResNet-50)	40.5	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
24	DeVIS (ResNet-50)	38	No	DeVIS: Making Deformable Transformers Work for V...	2022-07-22	Code
25	InstanceFormer (ResNet-50)	36.1	Yes	InstanceFormer: An Online Video Instance Segment...	2022-08-22	Code
26	STMask(R101-DCN-FPN)	29.4	No	Spatial Feature Calibration and Temporal Fusion ...	2021-04-06	Code

#1CAVIS(VIT-L, Offline)SOTA
49.7
AR1· Extra Data· 2024-07-03
Context-Aware Video Instance Segmentation Code
#2DVIS-DAQ(VIT-L, Offline)SOTA
49.6
AR1· Extra Data· 2024-03-29
DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries Code
#3DVIS++(VIT-L, Online)SOTA
49.5
AR1· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#4GenVIS (Swin-L)SOTA
49.1
AR1· Extra Data· 2022-11-16
A Generalized Framework for Video Instance Segmentation Code
#5DVIS++(VIT-L, Offline)
48.8
AR1· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#6GRAtt-VIS (Swin-L)
48.8
AR1· Extra Data· 2023-05-26
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation Code
#7RefineVIS (Swin-L, online)
48.3
AR1· Extra Data· 2023-06-07
RefineVIS: Video Instance Segmentation with Temporal Attention Refinement
#8NOVIS (Swin-L)
47.9
AR1· Extra Data· 2023-08-29
NOVIS: A Case for End-to-End Near-Online Video Instance Segmentation
#9DVIS(Swin-L)
47.7
AR1· Extra Data· 2023-06-06
DVIS: Decoupled Video Instance Segmentation Framework Code
#10VITA (Swin-L)SOTA
47.7
AR1· Extra Data· 2022-06-09
VITA: Video Instance Segmentation via Object Token Association Code
#11TarViS (Swin-L)
47.6
AR1· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#12Tube-Link(Swin-L)
47.5
AR1· 2023-03-22
Tube-Link: A Flexible Cross Tube Framework for Universal Video Segmentation Code
#13UniVS(Swin-L)
46.2
AR1· Extra Data· 2024-02-28
UniVS: Unified and Universal Video Segmentation with Prompts as Queries Code
#14MinVIS (Swin-L)
45.9
AR1· 2022-08-03
MinVIS: A Minimal Video Instance Segmentation Framework without Video-based Training Code
#15MDQE(Swin-L)
45.4
AR1· 2023-03-25
MDQE: Mining Discriminative Query Embeddings to Segment Occluded Instances on Challenging Videos Code
#16IDOL (Swin-L)
45
AR1· 2022-07-21
In Defense of Online Models for Video Instance Segmentation Code
#17BoxVIS(Swin-L & Box-sup)
44.8
AR1· 2023-03-26
BoxVIS: Video Instance Segmentation with Box Annotations Code
#18DeVIS (Swin-L)
43.8
AR1· 2022-07-22
DeVIS: Making Deformable Transformers Work for Video Instance Segmentation Code
#19InstanceFormer (Swin-L)
42.8
AR1· Extra Data· 2022-08-22
InstanceFormer: An Online Video Instance Segmentation Framework Code
#20TarViS (Swin-T)
42.2
AR1· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#21GRAtt-VIS (ResNet-50)
41.8
AR1· Extra Data· 2023-05-26
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation Code
#22NOVIS (ResNet-50)
41.3
AR1· Extra Data· 2023-08-29
NOVIS: A Case for End-to-End Near-Online Video Instance Segmentation
#23TarViS (ResNet-50)
40.5
AR1· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#24DeVIS (ResNet-50)
38
AR1· 2022-07-22
DeVIS: Making Deformable Transformers Work for Video Instance Segmentation Code
#25InstanceFormer (ResNet-50)
36.1
AR1· Extra Data· 2022-08-22
InstanceFormer: An Online Video Instance Segmentation Framework Code
#26STMask(R101-DCN-FPN)SOTA
29.4
AR1· 2021-04-06
Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation Code