[ICLR2021] (ViT) An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
ICLR2021
Link: [2010.11929] An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (arxiv.org)
Code: lucidrains/vit-pytorch: Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch (github.com)
1 Method
1.1 图像预处理
1.1.1 原始输入(Input)
输入图像尺寸224 x 224 x 3ÿ