# LiheYoung/Depth-Anything

[CVPR 2024] Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data. Foundation Model for Monocular Depth Estimation

Repository: https://github.com/LiheYoung/Depth-Anything
Canonical: https://ross.abutalabs.com/products/depth-anything
Homepage: https://depth-anything.github.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: depth-estimation, image-synthesis, metric-depth-estimation, monocular-depth-estimation
Last push: 2024-07-17T10:46:05+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 68
- inputs: {"age_days": 955, "days_push": 777, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8195, forks 618 (observed 2026-08-28T04:10:19.535751+00:00)

## What it is
Depth Anything is a monocular depth estimation foundation model trained on 1.5M labeled and 62M+ unlabeled images, released as a Python library with pretrained checkpoints. It supports zero-shot relative and metric depth estimation and integrates with ControlNet for image generation.

## Use cases
- estimate depth from a single image
- generate depth maps for photos
- use depth conditioning in stable diffusion controlnet
- run monocular depth estimation on video frames
- fine-tune a depth model on NYUv2 or KITTI
- get metric depth estimates from images

## When to choose
- you need robust zero-shot depth estimation from single images
- you want a depth processor for diffusion-based image generation
- you need a well-validated model with ONNX and TensorRT deployments available

## When to avoid
- you need real-time depth on edge devices without GPU acceleration
- you need multi-view or stereo depth from camera rigs
- you want the newest model quality - Depth Anything V2 supersedes this version

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing, deep-learning
- domain: computer-vision, deep-learning, artificial-intelligence, image-processing
- platform: python, cross-platform
- tags: depth-estimation, monocular-depth, foundation-model, controlnet, stable-diffusion, cvpr-2024, onnx, tensorrt, gpu

## Member repositories
- LiheYoung/Depth-Anything (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:19.535751+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 2026-08-29T17:29:33.836359+00:00, confidence not recorded.
  - readme: https://github.com/LiheYoung/Depth-Anything (fetched 2026-08-28T04:10:19.535751+00:00, sha 4a40378654ab)
  - homepage: https://depth-anything.github.io (fetched 2026-08-29T08:28:19.960480+00:00, sha d2f902917c35)
- Data as of 2026-08-30T08:39:29.467469+00:00.