HyperAIHyperAI超神经

Command Palette

Search for a command to run...

算力平台

Semi Supervised Semantic Segmentation On 21

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
AllSpark81.65AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation
PrevMatch (ResNet-101)81.4Revisiting and Maximizing Temporal Knowledge in Semi-supervised Semantic Segmentation
CorrMatch (Deeplabv3+ with ResNet-101)81.3CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation
UniMatch (DeepLab v3+ with ResNet-101)80.94Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation
S4MC78.84Semi-Supervised Semantic Segmentation via Marginal Contextual Information
Dual Teacher78.82Switching Temporary Teachers for Semi-Supervised Semantic Segmentation
PCR (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)78.6Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization
U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)77.21Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels
AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)77.2Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning
n-CPS (ResNet-101)75.86n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation
CPCL (DeepLab v3+ with ResNet-101)73.44Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation
n-CPS (ResNet-50)72.03n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation
CPCL (DeepLab v3+ with ResNet-50)71.66Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation
PCT (DeepLab v3+ with ResNet-50 pretrained on ImageNet-1K)71.35Learning Pseudo Labels for Semi-and-Weakly Supervised Semantic Segmentation
Dense FixMatch (DeepLabv3+ ResNet-101, over-sampling, single pass eval)54.85Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks
Dense FixMatch (DeepLabv3+ ResNet-50, over-sampling, single pass eval)52.15Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks
0 of 16 row(s) selected.
Semi Supervised Semantic Segmentation On 21 | SOTA | HyperAI超神经