Back to Reading Room

LA BIBLIOTHÈQUE · READING NOTE

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Dosovitskiy et al.

To Read · 想看

AIComputer VisionPapers

ViT, Vision Transformer, Image Patches, Transformer

Vision Transformer 论文。将图像切分成 Patch 并视作 Token 输入 Transformer,证明 Transformer 可以成为视觉领域的通用基础架构。