DINOv3 ViT-L/16 DPT Head

JSON →
meta vision
image

A monocular depth estimation model combining DINOv3 vision transformer features with a DPT decoder head.

vision
releasedApr 2024