Models / meta / DINOv3 ViT-L/16 DPT Head

DINOv3 ViT-L/16 DPT Head

JSON →
metavision

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

imagevision
homepage ↗
Specs
Context & limits
—
context window
—
max output
Lifecycle
Dates
released2024-04-01
Resources
homepagehttps://huggingface.co/facebook/dinov3-vitl16-chmv2-dpt-head ↗