search

TongkunGuan / CCD

[ICCV2023] Self-supervised Character-to-Character Distillation for Text Recognition
https://openaccess.thecvf.com/content/ICCV2023/papers/Guan_Self-Supervised_Character-to-Character_Distillation_for_Text_Recognition_ICCV_2023_paper.pdf
140 stars 7 forks source link

关于dataloader #12

Closed yjjexcellent closed 5 months ago

yjjexcellent commented 5 months ago

刚刚阅读这个代码，dataloader中datasetsupervised_kmeans返回的np.stack应该是不同增强视图的堆叠，请问在ABI DINOModel中，前向传播x1=[:,0]应该如何解释，我拿numpy简单抽象了一下，这个应该不是取两个增强视图拼接在一起吧

yjjexcellent commented 5 months ago

已解决