nianticlabs/monodepth2

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

4497 stars · 987 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Monodepth2 is the reference PyTorch implementation of the ICCV 2019 paper 'Digging into Self-Supervised Monocular Depth Prediction'. It trains and runs neural networks that estimate depth (disparity) from a single monocular image using self-supervision, with pretrained models provided.

Use cases

• estimate depth from a single image
• predict monocular disparity with a neural network
• train a self-supervised depth estimation model
• run monocular depth estimation on KITTI-style driving footage
• get metric depth from stereo-trained models
• reproduce ICCV 2019 monodepth2 research results

When to choose

• you need single-image depth estimation with pretrained PyTorch models
• you want to reproduce or build on the Monodepth2 paper
• you need a self-supervised depth prediction baseline for research

When to avoid

• you need real-time depth on mobile or embedded devices
• you need a maintained production library with modern PyTorch support
• you need commercial usage - the code is non-commercial only
• you need multi-frame or video depth estimation

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision image-processing computer-vision deep-learning machine-learning artificial-intelligence python depth-estimation monocular-depth self-supervised-learning pytorch disparity research-code iccv-2019 kitti linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="nianticlabs/monodepth2")