Domain Adaptation on GTA5-to-Cityscapes

Metric: mIoU (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	mIoU▼	Augmentations	Paper	Date↕	Code
1	CLUDA+HRDA	74.4	No	CLUDA : Contrastive Learning in Unsupervised Dom...	2022-08-27	Code
2	tqdm (EVA02-CLIP-L)	68.88	No	Textual Query-Driven Mask Transformer for Domain...	2024-07-12	Code
3	ADSI	67.75	No	Domain Generalization through Attenuation of Dom...	2025-04-09	Code
4	Rein	66.4	No	Stronger, Fewer, & Superior: Harnessing Vision F...	2023-12-07	Code
5	VLTSeg (EVA02-CLIP-L)	65.6	No	Strong but simple: A Baseline for Domain General...	2023-12-04	Code
6	DIFF	58.01	No	Diffusion Features to Bridge Domain Gap for Sema...	2024-06-02	Code
7	CMFormer	55.31	No	Learning Content-enhanced Mask Transformer for D...	2023-07-01	Code
8	Self-adaptation (ResNet - 101)	46.99	No	Semantic Self-adaptation: Enhancing Generalizati...	2022-08-10	Code
9	GtA-SFDA Source-Only (DeepLabv2-ResNet101)	43.5	No	Generalize then Adapt: Source-Free Domain Adapti...	2021-08-25	Code

#1CLUDA+HRDASOTA
74.4
mIoU· 2022-08-27
CLUDA : Contrastive Learning in Unsupervised Domain Adaptation for Semantic Segmentation Code
#2tqdm (EVA02-CLIP-L)
68.88
mIoU· 2024-07-12
Textual Query-Driven Mask Transformer for Domain Generalized Segmentation Code
#3ADSI
67.75
mIoU· 2025-04-09
Domain Generalization through Attenuation of Domain-Specific Information Code
#4Rein
66.4
mIoU· 2023-12-07
Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation Code
#5VLTSeg (EVA02-CLIP-L)
65.6
mIoU· 2023-12-04
Strong but simple: A Baseline for Domain Generalized Dense Perception by CLIP-based Transfer Learning Code
#6DIFF
58.01
mIoU· 2024-06-02
Diffusion Features to Bridge Domain Gap for Semantic Segmentation Code
#7CMFormer
55.31
mIoU· 2023-07-01
Learning Content-enhanced Mask Transformer for Domain Generalized Urban-Scene Segmentation Code
#8Self-adaptation (ResNet - 101)SOTA
46.99
mIoU· 2022-08-10
Semantic Self-adaptation: Enhancing Generalization with a Single Sample Code
#9GtA-SFDA Source-Only (DeepLabv2-ResNet101)SOTA
43.5
mIoU· 2021-08-25
Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation Code