swin-transformer

Swin Transformer是一种特为视觉领域设计的分层Transformer结构，它的两个主要特性是滑动窗口和分层表示。滑动窗口允许在局部不重叠的窗口中计算自注意力，并且允许跨窗口连接。分层表示允许模型适应不同尺度的图像，并且计算复杂度与图像大小成线性关系。因此，Swin Transformer被形象地称为"披着Transformer皮的CNN"。Swim Transformer可以用作通用的主干网络，适用于图像分类、图像分割、目标检测等多种计算机视觉任务。它以Vision Transformer(VIT)为基础，吸取了ResNet的精华，并通过逐步扩大感受野的设计思想来将Transformer应用于图像处理任务。Swin Transformer的成功并非偶然，它是基于丰富的经验和积累而来的。<span class="em">1</span><span class="em">2</span><span class="em">3</span>
#### 引用[.reference_title]
- *1* *2* [Swin-Transformer网络结构详解](https://blog.csdn.net/qq_37541097/article/details/121119988)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 50%"]
- *3* [神经网络学习笔记5——Swin-Transformer网络](https://blog.csdn.net/qq_45848817/article/details/127105956)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 50%"]
[ .reference_list ]