Swin Transformer：使用移位窗口的分层视觉Transformer【通过patch merging在不同stage逐步降低分辨率、提高通道数，输出一组“多尺度特征图”】【202103】

第001/14页（英文原文）Swin Transformer: Hierarchical Vision Transformer using Shifted WindowsZe Liu†* Yutong Lin†* Yue Cao* Han Hu*‡ Yixuan Wei† Zheng Zhang Stephen Lin Baining Guo Microsoft Research Asia{v-zeliu1,v-yutlin,yuecao,hanhu,v-yixwe,zhez,stevelin,bainguo}@microsoft.comAbstractThis paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities