Semantic Segmentation On Pascal Voc 2012

评估指标

Mean IoU

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
DeepLabv3+ (Xception-JFT)	89.0%	Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
DeepLabv3+ (Xception-65-JFT)	89.0%	Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
DeepLabv3-JFT	86.9%	Rethinking Atrous Convolution for Semantic Image Segmentation
CASIA_IVA_SDN	86.6%	Stacked Deconvolutional Network for Semantic Segmentation	-
Smooth Network with Channel Attention Block	86.2%	Learning a Discriminative Feature Network for Semantic Segmentation
SANet (pretraining on COCO dataset)	86.1%	Squeeze-and-Attention Networks for Semantic Segmentation
HamNet w/o COCO (ResNet-101)	85.9%	Is Attention Better Than Matrix Decomposition?
EncNet	85.9%	Context Encoding for Semantic Segmentation
Auto-DeepLab-L	85.6%	Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
PSPNet	85.4%	Pyramid Scene Parsing Network
ResNet-38 MS COCO	84.9%	Wider or Deeper: Revisiting the ResNet Model for Visual Recognition
OCR (HRNetV2-W48)	84.5%	Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation
OCR (ResNet-101)	84.3%	Segmentation Transformer: Object-Contextual Representations for Semantic Segmentation
Multipath-RefineNet	84.2%	RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation
ShelfNet	84.2%	ShelfNet for Fast Semantic Segmentation
EANet (ResNet-101)	84%	Beyond Self-attention: External Attention using Two Linear Layers for Visual Tasks
Large Kernel Matters	83.6%	Large Kernel Matters -- Improve Semantic Segmentation by Global Convolutional Network
TripleNet	83.3%	Triply Supervised Decoder Networks for Joint Detection and Segmentation	-
SANet	83.2%	Squeeze-and-Attention Networks for Semantic Segmentation
TuSimple	83.1%	Understanding Convolution for Semantic Segmentation

0 of 51 row(s) selected.