Video-Depth-Anything-Small

by bigshanedogg depth-estimation transformers apache-2.0

4 downloads 0 likes on Hugging Face updated 2026-10-27

Papers using or citing this model 1

Video Depth Anything: Consistent Depth Estimation for Super-Long Videos 282 citations · 2025 · arXiv
Depth Anything has achieved remarkable success in monocular depth estimation with strong generalization ability. However, it suffers from temporal inconsistency in videos, hindering its practical applications. Various methods have been...

Get this model

hp get bigshanedogg/Video-Depth-Anything-Small

The huggingpeers CLI is not released yet. Once it ships, the command above will download the model. Until then, download the weights from Hugging Face.

Details

HuggingPeers ID: ID hpm-r4Ni7pyWsCw
Provider: Hugging Face
Revision: fb8632eaa304
Organization: bigshanedogg
Pipeline: depth-estimation
Library: transformers
License: apache-2.0
Parameters: 29.1M
Published: 2026-10-27

Files 17

Total size: 111.1 MiB
LFS files: 1
Parameters: 29.1M
Revision: fb8632eaa304

Browse files & hashes →

Code

GitHub · DepthAnything/Video-Depth-Anything ↗

Linked from the model card — we do not mirror source repositories.

Sources

Hugging Face · bigshanedogg/Video-Depth-Anything-Small (this) · checked recently

View on Hugging Face ↗