HyperAI超神経

Semantic Segmentation On Isprs Potsdam

評価指標

Mean F1
Mean IoU
Overall Accuracy

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Mean F1
Mean IoU
Overall Accuracy
Paper TitleRepository
AerialFormer-B94.189.193.9AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
LSKNet-S93.187.292.0LSKNet: A Foundation Lightweight Backbone for Remote Sensing
PSPNet (SAP)-74.388.56Stochastic Subsampling With Average Pooling-
ViT-B + RVSA-UperNet--90.77Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
EfficientUNets and Transformers93.7-91.8Semantic Labeling of High Resolution Images Using EfficientUNets and Transformers-
BANet--91.06Transformer Meets Convolution: A Bilateral Awareness Network for Semantic Segmentation of Very Fine Resolution Urban Scene Images
FT-UNetFormer93.387.592.0UNetFormer: A UNet-like Transformer for Efficient Semantic Segmentation of Remote Sensing Urban Scene Imagery
RSP-Swin-T-UperNet--90.78An Empirical Study of Remote Sensing Pretraining
MANet--91.318Multiattention network for semantic segmentation of fine-resolution remote sensing images-
U-Net (ConvFormer-M36)-89.45-U-Net Ensemble for Enhanced Semantic Segmentation in Remote Sensing Imagery-
ViTAE-B + RVSA -UperNet--91.22Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
ViT-G12X492.12-92.58A Billion-scale Foundation Model for Remote Sensing Images-
RSP-ResNet-50-UperNet--90.61An Empirical Study of Remote Sensing Pretraining
RSP-ViTAEv2-S-UperNet--91.21An Empirical Study of Remote Sensing Pretraining
ABCNet--91.3ABCNet: Attentive Bilateral Contextual Network for Efficient Semantic Segmentation of Fine-Resolution Remote Sensing Images
IMP-ViTAEv2-S-UperNet--91.6An Empirical Study of Remote Sensing Pretraining
UNetFormer92.886.891.3UNetFormer: A UNet-like Transformer for Efficient Semantic Segmentation of Remote Sensing Urban Scene Imagery
DC-Swin93.2587.5692.0A Novel Transformer Based Semantic Segmentation Scheme for Fine-Resolution Remote Sensing Images
SFA-Net93.5--SFA-Net: Semantic Feature Adjustment Network for Remote Sensing Image Segmentation
0 of 19 row(s) selected.