HyperAI

Semantic Segmentation On Pascal Context

Metrics

mIoU

Results

Performance results of various models on this benchmark

Model Name
mIoU
Paper TitleRepository
EncNet (ResNet-101)51.7Context Encoding for Semantic Segmentation
HRNetV2 HRNetV2-W4854Deep High-Resolution Representation Learning for Visual Recognition
CondNet(ResNest-101)57CondNet: Conditional Classifier for Scene Segmentation
ViT-Adapter-L (Mask2Former, BEiT pretrain)68.2Vision Transformer Adapter for Dense Predictions
SegViT (ours)65.3SegViT: Semantic Segmentation with Plain Vision Transformers
ResNeSt-20058.4ResNeSt: Split-Attention Networks
OCR (HRNetV2-W48)56.2Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation
DGCNet (MS, ResNet-101)53.7Dual Graph Convolutional Network for Semantic Segmentation
CAA + Simple decoder (Efficientnet-B7)60.5Channelized Axial Attention for Semantic Segmentation -- Considering Channel Relation within Spatial Attention for Semantic Segmentation
LaU-regression-loss (ResNet-101)53.9Location-aware Upsampling for Semantic Segmentation
PlainSeg (EVA-02-L)71.0Minimalist and High-Performance Semantic Segmentation with Plain Vision Transformers
CAA + CAR (ConvNeXt-Large + JPU)64.1CAR: Class-aware Regularizations for Semantic Segmentation
CFM34.4Convolutional Feature Masking for Joint Object and Stuff Segmentation
DeepLabV245.7DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
InternImage-H70.3InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
Dilated-FCN2s42.6Efficient Yet Deep Convolutional Neural Networks for Semantic Segmentation
HRNetV2 + OCR + RMI (PaddleClas pretrained)59.6Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation
Asymmetric ALNN52.8Asymmetric Non-local Neural Networks for Semantic Segmentation
SenFormer (ResNet-101)56.6Efficient Self-Ensemble for Semantic Segmentation
VeryDeep44.5Bridging Category-level and Instance-level Semantic Image Segmentation-
0 of 66 row(s) selected.