Clarification on the training datasets used.

NVlabs / GroupViT

Official PyTorch implementation of GroupViT: Semantic Segmentation Emerges from Text Supervision, CVPR 2022.

https://arxiv.org/abs/2202.11094

Other

705 stars 53 forks source link

Clarification on the training datasets used. #46

Closed yash0307 closed 1 year ago

yash0307 commented 1 year ago

Hi, for the model that achieves 52.3% mIoU on PASCAL VOC, the paper says that GCC12M + YFCC14M are used for training. Whereas, in the config files of this repository GCC3M is also used. Which one is correct?

In case if GCC3M is also used, are the pre-trainings for fully supervised transfer in Table 5. done using GCC3M as well?

Thank you in advance.

xvjiarui commented 1 year ago

Hi @yash0307

Sorry for the confusion.

Yes. We are using CC3M as well.

yash0307 commented 1 year ago

Thank for the quick reply.