HyperAI
HomeNewsLatest PapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAI

Semantic Segmentation On Isaid

Metrics

mIoU

Results

Performance results of various models on this benchmark

Model Name
mIoU
Paper TitleRepository
ViTAE-B + RVSA-UperNet64.49Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
FarSeg@ResNet-5063.71Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery
AerialFormer-S68.4AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
SegNeXt-L70.3SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
FarSeg++@Swin-T66.3FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery
RSP-ResNet-50-UperNet61.6An Empirical Study of Remote Sensing Pretraining
SegNeXt-S68.8SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
DeepLabV3 with R-5067.03Resolution-Aware Design of Atrous Rates for Semantic Segmentation Networks-
SegNeXt-B69.9SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
SegNeXt-T68.3SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
ViT-B + RVSA-UperNet63.85Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
IMP-ViTAEv2-S-UperNet65.3An Empirical Study of Remote Sensing Pretraining
FarSeg++@MiT-B267.9FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery
FarSeg++@ResNet-5067.6FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery
FactSeg@ResNet-5064.79FactSeg: Foreground Activation Driven Small Object Semantic Segmentation in Large-Scale Remote Sensing Imagery
RSP-Swin-T-UperNet64.1An Empirical Study of Remote Sensing Pretraining
AerialFormer-T67.5AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
AerialFormer-B69.3AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation
RSP-ViTAEv2-S-UperNet64.3An Empirical Study of Remote Sensing Pretraining
0 of 19 row(s) selected.