Unsupervised Domain Adaptation On Synthia To

評価指標

mIoU

mIoU (13 classes)

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	mIoU	mIoU (13 classes)	Paper Title	Repository
Re-EnD-UDA	52.58	59.95	Rethinking Ensemble-Distillation for Semantic Segmentation Based Unsupervised Domain Adaptation
DAFormer + ProCST	-	68.2	ProCST: Boosting Semantic Segmentation Using Progressive Cyclic Style-Transfer
DCF	69.3	75.9	Transferring to Real-World Layouts: A Depth-aware Framework for Scene Adaptation
MRNet + Adaboost	45.9	52.9	Adaptive Boosting for Domain Adaptation: Towards Robust Predictions in Scene Segmentation
Uncertainty	47.9	54.9	Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation
Uncertainty + Adaboost	50.4	57.5	Adaptive Boosting for Domain Adaptation: Towards Robust Predictions in Scene Segmentation
HRDA	65.8	72.4	HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation
MIC	67.3	74.0	MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation
DIDA	63.3	70.1	Dual-level Interaction for Domain Adaptive Semantic Segmentation
CLUDA+HRDA	67.2	-	CLUDA : Contrastive Learning in Unsupervised Domain Adaptation for Semantic Segmentation
MRNet	43.2	50.2	Unsupervised Scene Adaptation with Memory Regularization in vivo
SePiCo (DeepLabv2 ResNet-101)	-	66.5	SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation	-
TransDA-B	-	66.3	Smoothing Matters: Momentum Transformer for Domain Adaptive Semantic Segmentation
DAFormer+CSI	61.4	-	Cross-Domain Semantic Segmentation on Inconsistent Taxonomy using VLMs
ProCA(ResNet-101)	-	59.6	Prototypical Contrast Adaptation for Domain Adaptive Semantic Segmentation
DAFormer	60.9	67.4	DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation
HRDA + PiPa	-	74.8	PiPa: Pixel- and Patch-wise Self-supervised Learning for Domain Adaptative Semantic Segmentation
FAFS	54.5	61.4	A Novel Unsupervised Domain Adaption Method for Depth-Guided Semantic Segmentation Using Coarse-to-Fine Alignment	-
MIC+CSI	68.8	75.9	Cross-Domain Semantic Segmentation on Inconsistent Taxonomy using VLMs
SePiCo	-	71.4	SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation	-

0 of 21 row(s) selected.