Semi Supervised Semantic Segmentation On 9

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 TitleRepository
Dual Teacher81.03Switching Temporary Teachers for Semi-Supervised Semantic Segmentation-
AllSpark80.92AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation
CorrMatch (Deeplabv3+ with ResNet-101)80.9CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation
PrevMatch (ResNet-101)80.8Revisiting and Maximizing Temporal Knowledge in Semi-supervised Semantic Segmentation
PCR (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)80.78Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization
UniMatch (DeepLab v3+ with ResNet-101)80.43Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation
S4MC79.85Semi-Supervised Semantic Segmentation via Marginal Contextual Information
U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)79.3Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels
n-CPS (ResNet-101)78.97n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation-
PS-MT (DeepLab v3+ with ImageNet-pretrained ResNet-101, single scale inference)78.72Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation
AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)78.06Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning
GuidedMix-Net(DeepLab v2 with ResNet101, input-size: 512x512 with multi-scale and flip, ImageNet pretrained)77.8%GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference
CPS (DeepLab v3+ with ImageNet-pretrained ResNet-101, single scale inference)77.68%Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
CPCL (DeepLab v3+ with ResNet-101)77.16Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation
Error Localization Network (DeeplabV3 with ResNet-101)76.58%Semi-supervised Semantic Segmentation with Error Localization Network
PCT (DeepLab v3+ with ResNet-50 pretrained on ImageNet-1K)76.47Learning Pseudo Labels for Semi-and-Weakly Supervised Semantic Segmentation-
CW-BASS (DeepLab v3+ with ResNet-50)76.2CW-BASS: Confidence-Weighted Boundary-Aware Learning for Semi-Supervised Semantic Segmentation
n-CPS (ResNet-50)75.85n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation-
GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)75.5%GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference
Error Localization Network (DeeplabV3 with ResNet-50)74.63%Semi-supervised Semantic Segmentation with Error Localization Network
0 of 24 row(s) selected.
Semi Supervised Semantic Segmentation On 9 | SOTA | HyperAI超神经