Share pretrained DINOv2 weights of ViT on ImageNet-22k

facebookresearch / dinov2

PyTorch code and models for the DINOv2 self-supervised learning method.

Apache License 2.0

9.16k stars 817 forks source link

Share pretrained DINOv2 weights of ViT on ImageNet-22k #287

Open black0017 opened 1 year ago

black0017 commented 1 year ago

Hi all, and thank you for the amazing work!

Is there any chance you could share the weights/checkpoints of DINOv2 ViT-L (Table 1 in the paper) or ViT-H (used for Self-supervised image retrieval in the paper) pretrained ONLY on ImageNet-22K (aka 21K)?

That could be amazing!

Best, N.A.

black0017 commented 7 months ago

Maybe @piotr-bojanowski ?