| Benchmark | Model | Rank | Results |
|---|---|---|---|
| semantic-segmentation-on-cityscapes-val | PatchDiverse + Swin-L (multi-scale test, upernet, ImageNet22k pretrain) | #22 | mIoU: 83.6% |
| semantic-segmentation-on-ade20k-val | PatchDiverse + Swin-L (multi-scale test, upernet, ImageNet22k pretrain) | #31 | mIoU: 54.4% |
| semantic-segmentation-on-ade20k | PatchDiverse + Swin-L (multi-scale test, upernet, ImageNet22k pretrain) | #62 | Validation mIoU: 54.4 |