OpenLRM icon indicating copy to clipboard operation
OpenLRM copied to clipboard

DINO encoder

Open mli0603 opened this issue 11 months ago • 1 comments

Dear authors:

Thank you for your great work! I wonder why you used DINOV1 instead DINOV2, which is more suitable for dense prediction task. Thank you!

mli0603 avatar Mar 05 '24 05:03 mli0603

Hi,

Thanks for your interest!

We used DINOv1 during our first development to align with the original LRM paper and to avoid introducing new factors. Now that the codebase is been verified in v1.0, we switch to DINOv2 to ensure its absolute performance.

ZexinHe avatar Mar 05 '24 05:03 ZexinHe