Semi Supervised Semantic Segmentation On 4

评估指标

Validation mIoU

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
AllSpark	82.04%	AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation
UniMatch	81.92%	Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation
PrevMatch (ResNet-101)	81.9	Revisiting and Maximizing Temporal Knowledge in Semi-supervised Semantic Segmentation
CorrMatch (Deeplabv3+ with ResNet-101)	81.9%	CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation
Dual Teacher	81.19	Switching Temporary Teachers for Semi-Supervised Semantic Segmentation	-
PCR (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)	80.71%	Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization
S4MC	79.67%	Semi-Supervised Semantic Segmentation via Marginal Contextual Information
U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)	79.01%	Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels
PS-MT	78.20%	Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation
n-CPS (ResNet-101)	77.99%	n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation	-
AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)	77.57%	Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning
CPS	76.44%	Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
GuidedMix-Net	76.4%	GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference
CPCL (DeepLab v3+ with ResNet-101)	76.4%	Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation
CW-BASS (DeepLab v3+ with ResNet-50)	75.81%	CW-BASS: Confidence-Weighted Boundary-Aware Learning for Semi-Supervised Semantic Segmentation
PS-MT	75.70%	Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation
PCT (DeepLab v3+ with ResNet-50 pretrained on ImageNet-1K)	75.52%	Learning Pseudo Labels for Semi-and-Weakly Supervised Semantic Segmentation	-
Error Localization Network	75.10%	Semi-supervised Semantic Segmentation with Error Localization Network
n-CPS	74.21%	n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation	-
CPCL (DeepLab v3+ with ResNet-50)	73.74%	Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation

0 of 35 row(s) selected.