Semantic Segmentation on Cityscapes (val) (Single/Multi Scale mIoU)
83.9mIoU (Single Scale)DINAT-L
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DINAT-LBackbone=DINAT-L, Win. Size=11 x 11, # of Params=220 M, FLOPs=509 G, Pre-trained=ImageNet-22K, Resolution=800x800, Framework=Mask2Former2022.09 | 83.9 | 84.5 | |
| Swin-LBackbone=Swin-L, Win. Size=12 x 12, # of Params=215 M, FLOPs=627 G, Pre-trained=ImageNet-22K, Resolution=800x800, Framework=Mask2Former2022.09 | 83.3 | 84.3 |