# lpiccinelli-eth/UniDepth

Universal Monocular Metric Depth Estimation

Repository: https://github.com/lpiccinelli-eth/UniDepth
Canonical: https://ross.abutalabs.com/products/unidepth
Language: Python
License: NOASSERTION
License Family: other
Topics: 3d-reconstruction, computer-vision, depth-estimation
Last push: 2025-05-18T21:52:52+00:00

## Health v2 (maintenance only)
Score: 35/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 22, release rhythm 35, longevity 63
- inputs: {"age_days": 891, "days_push": 472, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1246, forks 117 (observed 2026-08-28T04:04:07.200037+00:00)

## What it is
UniDepth is a Python library and research codebase for universal monocular metric depth estimation from single images, based on CVPR 2024 and UniDepthV2 papers from ETH Zurich. It provides pretrained models, inference, and training code that predict metric depth without requiring camera intrinsics.

## Use cases
- estimate metric depth from a single image
- 3d reconstruction from monocular photos
- depth estimation for robotics or autonomous navigation
- run state-of-the-art depth models on KITTI or NYU Depth
- get point clouds from single camera images

## When to choose
- you need metric (not just relative) depth from a single image without camera intrinsics
- you want a research-grade model with pretrained weights and training code
- you work in PyTorch and need benchmark-leading depth estimation

## When to avoid
- you need real-time depth on low-power or CPU-only hardware
- you need a non-research license for commercial use (license is custom)
- you need stereo or multi-view depth rather than monocular

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, machine-learning, artificial-intelligence, robotics
- platform: python
- tags: depth-estimation, monocular-depth, metric-depth, 3d-reconstruction, research-code, pytorch, linux, gpu

## Member repositories
- lpiccinelli-eth/UniDepth (main) score 35

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:07.200037+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:08:04.763904+00:00, confidence not recorded.
  - readme: https://github.com/lpiccinelli-eth/UniDepth (fetched 2026-08-28T04:04:07.200037+00:00, sha 174e8b4c03e7)
- Data as of 2026-08-30T08:39:29.467469+00:00.