Domain Adaptation On Gta5 To Cityscapes

Metriken

mIoU

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	mIoU	Paper Title	Repository
HRDA+PiPa	75.6	PiPa: Pixel- and Patch-wise Self-supervised Learning for Domain Adaptative Semantic Segmentation
DAFormer	68.3	DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation
ProDA	57.5	Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation
TransDA-B	63.9	Smoothing Matters: Momentum Transformer for Domain Adaptive Semantic Segmentation
FREDOM - Transformer	73.6	FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding
ILM-ASSL	76.1	Iterative Loop Method Combining Active and Semi-Supervised Learning for Domain Adaptive Semantic Segmentation
MIC	75.9	MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation
FAFS	58.8	A Novel Unsupervised Domain Adaption Method for Depth-Guided Semantic Segmentation Using Coarse-to-Fine Alignment	-
SePiCo	70.3	SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation	-
HALO	77.8	Hyperbolic Active Learning for Semantic Segmentation under Domain Shift
ProDA+CRA	58.6	Cross-Region Domain Adaptation for Class-level Alignment	-
FREDOM - DeepLabV2	61.3	FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding
DCF	77.7	Transferring to Real-World Layouts: A Depth-aware Framework for Scene Adaptation
GtA-SFDA (DeepLabv2-ResNet101)	53.4	Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation
DAFormer + ProCST	69.4	ProCST: Boosting Semantic Segmentation Using Progressive Cyclic Style-Transfer
BiSMAP	61.2	Bidirectional Self-Training with Multiple Anisotropic Prototypes for Domain Adaptive Semantic Segmentation
G2L	59.7	G2L: A Global to Local Alignment Method for Unsupervised Domain Adaptive Semantic Segmentation	-
MRNet	48.3	Unsupervised Scene Adaptation with Memory Regularization in vivo
EHTDI*	62.0	Exploring High-quality Target Domain Information for Unsupervised Domain Adaptive Semantic Segmentation
MRNet + Rectifying Label	50.3	Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation

0 of 26 row(s) selected.