VPNeXt -- Rethinking Dense Decoding for Plain Vision Transformer

Open paper

Benchmark	Model	Rank	Results
semantic-segmentation-on-cityscapes-val	VPNeXt	–	mIoU: 84.4
semantic-segmentation-on-coco-stuff-test	VPNeXt	–	mIoU: 53.7
semantic-segmentation-on-pascal-context	VPNeXt	–	mIoU: 71.1