Semantic Segmentation On Loveda

Metrics

Category mIoU

Results

Performance results of various models on this benchmark

		Paper Title
U-Net (MaxViT-S)	56.16	U-Net Ensemble for Enhanced Semantic Segmentation in Remote Sensing Imagery
SFA-Net	54.9	SFA-Net: Semantic Feature Adjustment Network for Remote Sensing Image Segmentation
ViT-G12X4	54.4	A Billion-scale Foundation Model for Remote Sensing Images
SelectiveMAE+ViT-L	54.31	Scaling Efficient Masked Image Modeling on Large Remote Sensing Dataset
MAE+MTP(ViT-L+RVSA)	54.17	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
IMP+MTP(InternImage-XL)	54.17	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
AerialFormer-B	54.1	AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
LSKNet-S	54.0	Large Selective Kernel Network for Remote Sensing Object Detection
LWGANet L2	53.6	LWGANet: A Lightweight Group Attention Backbone for Remote Sensing Visual Tasks
LOGCAN++	53.35	LOGCAN++: Adaptive Local-global class-aware network for semantic segmentation of remote sensing imagery
LSKNet-T	53.2	Large Selective Kernel Network for Remote Sensing Object Detection
DecoupleNet D2	53.1	DecoupleNet: A Lightweight Backbone Network With Efficient Feature Decoupling for Remote Sensing Visual Tasks
Hi-ResNet	52.6	Hi-ResNet: Edge Detail Enhancement for High-Resolution Remote Sensing Segmentation
ViTAE-B + RVSA-UperNet	52.44	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
UNetFormer	52.40	UNetFormer: A UNet-like Transformer for Efficient Semantic Segmentation of Remote Sensing Urban Scene Imagery
MAE+MTP(ViT-B+RVSA)	52.39	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
ViT-B + RVSA-UperNet	51.95	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
HRNetw32	49.79	LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation

0 of 18 row(s) selected.

Semantic Segmentation On Loveda

Metrics

Category mIoU

Results

Performance results of various models on this benchmark

		Paper Title
U-Net (MaxViT-S)	56.16	U-Net Ensemble for Enhanced Semantic Segmentation in Remote Sensing Imagery
SFA-Net	54.9	SFA-Net: Semantic Feature Adjustment Network for Remote Sensing Image Segmentation
ViT-G12X4	54.4	A Billion-scale Foundation Model for Remote Sensing Images
SelectiveMAE+ViT-L	54.31	Scaling Efficient Masked Image Modeling on Large Remote Sensing Dataset
MAE+MTP(ViT-L+RVSA)	54.17	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
IMP+MTP(InternImage-XL)	54.17	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
AerialFormer-B	54.1	AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
LSKNet-S	54.0	Large Selective Kernel Network for Remote Sensing Object Detection
LWGANet L2	53.6	LWGANet: A Lightweight Group Attention Backbone for Remote Sensing Visual Tasks
LOGCAN++	53.35	LOGCAN++: Adaptive Local-global class-aware network for semantic segmentation of remote sensing imagery
LSKNet-T	53.2	Large Selective Kernel Network for Remote Sensing Object Detection
DecoupleNet D2	53.1	DecoupleNet: A Lightweight Backbone Network With Efficient Feature Decoupling for Remote Sensing Visual Tasks
Hi-ResNet	52.6	Hi-ResNet: Edge Detail Enhancement for High-Resolution Remote Sensing Segmentation
ViTAE-B + RVSA-UperNet	52.44	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
UNetFormer	52.40	UNetFormer: A UNet-like Transformer for Efficient Semantic Segmentation of Remote Sensing Urban Scene Imagery
MAE+MTP(ViT-B+RVSA)	52.39	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
ViT-B + RVSA-UperNet	51.95	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
HRNetw32	49.79	LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation

0 of 18 row(s) selected.