Hi,
thank you for your great work!
For the downstream segmentation task it is stated in the paper that you use Mask2Former and ViT Adapter to make dense predictions.
How can I connect the timm checkpoint as a backbone together with the mask2former segmentation head? When trying to load your weights into the ViT Adapter, I receive several errors.
Any help is appreciated.
Hi, thank you for your great work! For the downstream segmentation task it is stated in the paper that you use Mask2Former and ViT Adapter to make dense predictions. How can I connect the timm checkpoint as a backbone together with the mask2former segmentation head? When trying to load your weights into the ViT Adapter, I receive several errors. Any help is appreciated.
Thank you and best, Leon