search

JDAI-CV / FaceX-Zoo

A PyTorch Toolbox for Face Recognition
https://arxiv.org/pdf/2101.04407.pdf
Other
1.88k stars 434 forks source link

SwinTransformer input size #116

Open otroshi opened 2 years ago

otroshi commented 2 years ago

Hi,

Do SwinTransformer based models get the 112 x 112 images as their input (like all other models)? or do they get 224 x 224 images?

PS. While in Step1: Prepare the training data images are mentioned to be 112 x 112, there is an argument in training_mode/backbone_conf.yaml and also model logs for img_size which is set 224

wang21jun commented 2 years ago

In order to stay consistent with the original setting of SwinTransformer, we just resize the input image from 112112 to 224224.