Fetching the paper…

Lawin Transformer: Improving Semantic Segmentation Transformer with Multi-Scale Representations via Large Window Attention · Around