Video Instance Segmentation on OVIS validation

Metric: AP50 (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	AP50▼	Extra Data	Paper	Date↕	Code
1	DVIS-DAQ(VIT-L, Offline)	83.8	Yes	DVIS-DAQ: Improving Video Segmentation via Dynam...	2024-03-29	Code
2	CAVIS(VIT-L, Offline)	82.6	Yes	Context-Aware Video Instance Segmentation	2024-07-03	Code
3	DVIS++(VIT-L,Offline)	78.9	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
4	DVIS(Swin-L, Offline)	75.9	No	DVIS: Decoupled Video Instance Segmentation Fram...	2023-06-06	Code
5	DVIS++(VIT-L, Online)	72.5	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
6	UNINEXT (ViT-H, Online)	72.5	Yes	Universal Instance Perception as Object Discover...	2023-03-12	Code
7	DVIS(Swin-L, Online)	71.9	No	DVIS: Decoupled Video Instance Segmentation Fram...	2023-06-06	Code
8	CTVIS (Swin-L)	71.5	Yes	CTVIS: Consistent Training for Online Video Inst...	2023-07-24	Code
9	RefineVIS (Swin-L, offline)	70.4	Yes	RefineVIS: Video Instance Segmentation with Temp...	2023-06-07	-
10	GenVIS (Swin-L)	69.2	Yes	A Generalized Framework for Video Instance Segme...	2022-11-16	Code
11	GRAtt-VIS (Swin-L)	69.1	Yes	GRAtt-VIS: Gated Residual Attention for Auto Rec...	2023-05-26	Code
12	DVIS++(R50, Offline)	68.9	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
13	BoxVIS(Swin-L & Box-sup)	68.4	No	BoxVIS: Video Instance Segmentation with Box Ann...	2023-03-26	Code
14	NOVIS (Swin-L)	68.3	Yes	NOVIS: A Case for End-to-End Near-Online Video I...	2023-08-29	-
15	TarViS (Swin-L)	67.8	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
16	MDQE(SwinL)	67.8	No	MDQE: Mining Discriminative Query Embeddings to ...	2023-03-25	Code
17	IDOL (Swin-L)	65.7	No	In Defense of Online Models for Video Instance S...	2022-07-21	Code
18	ROVIS (Swin-L)	64.7	No	Robust Online Video Instance Segmentation with T...	2022-11-16	Code
19	DVIS++(R50, Online)	62.8	Yes	DVIS++: Improved Decoupled Framework for Univers...	2023-12-20	Code
20	MinVIS (Swin-L)	61.5	No	MinVIS: A Minimal Video Instance Segmentation Fr...	2022-08-03	Code
21	GRAtt-VIS (ResNet-50)	60.8	Yes	GRAtt-VIS: Gated Residual Attention for Auto Rec...	2023-05-26	Code
22	CTVIS (ResNet-50)	60.8	Yes	CTVIS: Consistent Training for Online Video Inst...	2023-07-24	Code
23	DeVIS (Swin-L)	59.3	No	DeVIS: Making Deformable Transformers Work for V...	2022-07-22	Code
24	NOVIS (ResNet-50)	56.2	Yes	NOVIS: A Case for End-to-End Near-Online Video I...	2023-08-29	-
25	UNINEXT (ResNet-50, Online)	55.5	Yes	Universal Instance Perception as Object Discover...	2023-03-12	Code
26	TarViS (Swin-T)	55	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
27	TarViS (ResNet-50)	52.5	Yes	TarViS: A Unified Approach for Target-based Vide...	2023-01-06	Code
28	VITA (Swin-L)	51.9	Yes	VITA: Video Instance Segmentation via Object Tok...	2022-06-09	Code
29	Tube-Link(ResNet-50)	51.5	No	Tube-Link: A Flexible Cross Tube Framework for U...	2023-03-22	Code
30	IDOL (ResNet-50)	51.3	No	In Defense of Online Models for Video Instance S...	2022-07-21	Code
31	DeVIS (ResNet-50)	47.6	No	DeVIS: Making Deformable Transformers Work for V...	2022-07-22	Code
32	InstanceFormer (Swin-L)	42.5	Yes	InstanceFormer: An Online Video Instance Segment...	2022-08-22	Code
33	InstanceFormer(ResNet-50)	40.7	Yes	InstanceFormer: An Online Video Instance Segment...	2022-08-22	Code
34	Mask2Former-VIS	36.9	No	Mask2Former for Video Instance Segmentation	2021-12-20	Code
35	CrossVIS (ResNet-50, calibration)	35.5	No	Crossover Learning for Fast Online Video Instanc...	2021-04-13	Code
36	STMask(R101-DCN-FPN)	35.4	No	Spatial Feature Calibration and Temporal Fusion ...	2021-04-06	Code
37	TeViT (ResNet-50)	34.9	No	Temporally Efficient Vision Transformer for Vide...	2022-04-18	Code
38	CMaskTrack R-CNN (ResNet-50)	33.9	No	Occluded Video Instance Segmentation: A Benchmark	2021-02-02	Code
39	D2Conv3D (ResNet-50)	33.8	No	-	-	Code
40	STC (ResNet-50)	33.5	No	STC: Spatio-Temporal Contrastive Learning for Vi...	2022-02-08	-
41	CrossVIS (ResNet-50)	32.7	No	Crossover Learning for Fast Online Video Instanc...	2021-04-13	Code
42	CSipMask (ResNet-50)	29.9	No	Occluded Video Instance Segmentation: A Benchmark	2021-02-02	Code

#1DVIS-DAQ(VIT-L, Offline)SOTA
83.8
AP50· Extra Data· 2024-03-29
DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries Code
#2CAVIS(VIT-L, Offline)
82.6
AP50· Extra Data· 2024-07-03
Context-Aware Video Instance Segmentation Code
#3DVIS++(VIT-L,Offline)SOTA
78.9
AP50· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#4DVIS(Swin-L, Offline)SOTA
75.9
AP50· 2023-06-06
DVIS: Decoupled Video Instance Segmentation Framework Code
#5DVIS++(VIT-L, Online)
72.5
AP50· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#6UNINEXT (ViT-H, Online)SOTA
72.5
AP50· Extra Data· 2023-03-12
Universal Instance Perception as Object Discovery and Retrieval Code
#7DVIS(Swin-L, Online)
71.9
AP50· 2023-06-06
DVIS: Decoupled Video Instance Segmentation Framework Code
#8CTVIS (Swin-L)
71.5
AP50· Extra Data· 2023-07-24
CTVIS: Consistent Training for Online Video Instance Segmentation Code
#9RefineVIS (Swin-L, offline)
70.4
AP50· Extra Data· 2023-06-07
RefineVIS: Video Instance Segmentation with Temporal Attention Refinement
#10GenVIS (Swin-L)SOTA
69.2
AP50· Extra Data· 2022-11-16
A Generalized Framework for Video Instance Segmentation Code
#11GRAtt-VIS (Swin-L)
69.1
AP50· Extra Data· 2023-05-26
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation Code
#12DVIS++(R50, Offline)
68.9
AP50· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#13BoxVIS(Swin-L & Box-sup)
68.4
AP50· 2023-03-26
BoxVIS: Video Instance Segmentation with Box Annotations Code
#14NOVIS (Swin-L)
68.3
AP50· Extra Data· 2023-08-29
NOVIS: A Case for End-to-End Near-Online Video Instance Segmentation
#15TarViS (Swin-L)
67.8
AP50· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#16MDQE(SwinL)
67.8
AP50· 2023-03-25
MDQE: Mining Discriminative Query Embeddings to Segment Occluded Instances on Challenging Videos Code
#17IDOL (Swin-L)SOTA
65.7
AP50· 2022-07-21
In Defense of Online Models for Video Instance Segmentation Code
#18ROVIS (Swin-L)
64.7
AP50· 2022-11-16
Robust Online Video Instance Segmentation with Track Queries Code
#19DVIS++(R50, Online)
62.8
AP50· Extra Data· 2023-12-20
DVIS++: Improved Decoupled Framework for Universal Video Segmentation Code
#20MinVIS (Swin-L)
61.5
AP50· 2022-08-03
MinVIS: A Minimal Video Instance Segmentation Framework without Video-based Training Code
#21GRAtt-VIS (ResNet-50)
60.8
AP50· Extra Data· 2023-05-26
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation Code
#22CTVIS (ResNet-50)
60.8
AP50· Extra Data· 2023-07-24
CTVIS: Consistent Training for Online Video Instance Segmentation Code
#23DeVIS (Swin-L)
59.3
AP50· 2022-07-22
DeVIS: Making Deformable Transformers Work for Video Instance Segmentation Code
#24NOVIS (ResNet-50)
56.2
AP50· Extra Data· 2023-08-29
NOVIS: A Case for End-to-End Near-Online Video Instance Segmentation
#25UNINEXT (ResNet-50, Online)
55.5
AP50· Extra Data· 2023-03-12
Universal Instance Perception as Object Discovery and Retrieval Code
#26TarViS (Swin-T)
55
AP50· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#27TarViS (ResNet-50)
52.5
AP50· Extra Data· 2023-01-06
TarViS: A Unified Approach for Target-based Video Segmentation Code
#28VITA (Swin-L)SOTA
51.9
AP50· Extra Data· 2022-06-09
VITA: Video Instance Segmentation via Object Token Association Code
#29Tube-Link(ResNet-50)
51.5
AP50· 2023-03-22
Tube-Link: A Flexible Cross Tube Framework for Universal Video Segmentation Code
#30IDOL (ResNet-50)
51.3
AP50· 2022-07-21
In Defense of Online Models for Video Instance Segmentation Code
#31DeVIS (ResNet-50)
47.6
AP50· 2022-07-22
DeVIS: Making Deformable Transformers Work for Video Instance Segmentation Code
#32InstanceFormer (Swin-L)
42.5
AP50· Extra Data· 2022-08-22
InstanceFormer: An Online Video Instance Segmentation Framework Code
#33InstanceFormer(ResNet-50)
40.7
AP50· Extra Data· 2022-08-22
InstanceFormer: An Online Video Instance Segmentation Framework Code
#34Mask2Former-VISSOTA
36.9
AP50· 2021-12-20
Mask2Former for Video Instance Segmentation Code
#35CrossVIS (ResNet-50, calibration)SOTA
35.5
AP50· 2021-04-13
Crossover Learning for Fast Online Video Instance Segmentation Code
#36STMask(R101-DCN-FPN)SOTA
35.4
AP50· 2021-04-06
Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation Code
#37TeViT (ResNet-50)
34.9
AP50· 2022-04-18
Temporally Efficient Vision Transformer for Video Instance Segmentation Code
#38CMaskTrack R-CNN (ResNet-50)SOTA
33.9
AP50· 2021-02-02
Occluded Video Instance Segmentation: A Benchmark Code
#39D2Conv3D (ResNet-50)
33.8
AP50
No paperCode
#40STC (ResNet-50)
33.5
AP50· 2022-02-08
STC: Spatio-Temporal Contrastive Learning for Video Instance Segmentation
#41CrossVIS (ResNet-50)
32.7
AP50· 2021-04-13
Crossover Learning for Fast Online Video Instance Segmentation Code
#42CSipMask (ResNet-50)
29.9
AP50· 2021-02-02
Occluded Video Instance Segmentation: A Benchmark Code