Thank you for the great work on Fast3R!
In the paper, you mentioned the following:
"We follow DUSt3R’s design and use CroCo ViT as our encoder, though we found DINOv2 works similarly."
I was wondering — would it be possible for you to release the pretrained checkpoint using DINOv2 as the ViT encoder?
Thanks in advance!
Thank you for the great work on Fast3R!
In the paper, you mentioned the following:
"We follow DUSt3R’s design and use CroCo ViT as our encoder, though we found DINOv2 works similarly."
I was wondering — would it be possible for you to release the pretrained checkpoint using DINOv2 as the ViT encoder?
Thanks in advance!