Image To Image Translation On Synthia To

Metrics

mIoU (13 classes)

Results

Performance results of various models on this benchmark

Model Name	mIoU (13 classes)	Paper Title
DAFormer	67.4	DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation
CAMix (w Deeplabv2 ResNet 101)	59.7	Context-Aware Mixup for Domain Adaptive Semantic Segmentation
HRDA	72.4	HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation
LRENT (DeepLabv2)	48.7	Confidence Regularized Self-Training
Discriminative Patch (ResNet-101)	46.5	Domain Adaptation for Structured Output via Discriminative Patch Representations
ADVENT	48	ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation
IAST(ResNet-101)	57.0	Instance Adaptive Self-Training for Unsupervised Domain Adaptation
Multi-level Adaptation	46.7	Learning to Adapt Structured Output Space for Semantic Segmentation
PyCDA (ResNet-101)	53.3	Constructing Self-motivated Pyramid Curriculums for Cross-Domain Semantic Segmentation: A Non-Adversarial Approach
superpixel + color constancy	29.7	A Curriculum Domain Adaptation Approach to the Semantic Segmentation of Urban Scenes
MIC	74.0	MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation
CDA	29.0	Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes
FADA (ResNet-101)	52.5	Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation
CAG-UDA	44.5	Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation
DAFormer + ProCST	68.2	ProCST: Boosting Semantic Segmentation Using Progressive Cyclic Style-Transfer
TransDA-B	66.3	Smoothing Matters: Momentum Transformer for Domain Adaptive Semantic Segmentation
Domain Invariant Structure Extraction	41.5	All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation
HRDA + PiPa	74.8	PiPa: Pixel- and Patch-wise Self-supervised Learning for Domain Adaptative Semantic Segmentation
CPSL	65.3	Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
DADA (ResNet-101)	49.8	DADA: Depth-aware Domain Adaptation in Semantic Segmentation

0 of 28 row(s) selected.