# JiaRenChang/PSMNet

Pyramid Stereo Matching Network (CVPR2018)

Repository: https://github.com/JiaRenChang/PSMNet
Canonical: https://ross.abutalabs.com/products/psmnet
Language: Python
License: MIT
License Family: permissive
Topics: psmnet, stereo-matching, stereo-vision, pytorch
Last push: 2021-09-22T09:09:08+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3089, "days_push": 1806, "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 1548, forks 424 (observed 2026-08-28T04:05:01.826854+00:00)

## What it is
PSMNet is the official PyTorch implementation of the CVPR 2018 paper 'Pyramid Stereo Matching Network' for stereo depth estimation. It uses spatial pyramid pooling and a 3D CNN with stacked hourglass networks to compute disparity maps from stereo image pairs.

## Use cases
- estimate depth from stereo image pairs
- compute disparity maps with a CNN
- train a stereo matching model on Scene Flow
- evaluate stereo models on KITTI
- reproduce PSMNet research results
- benchmark stereo matching networks

## When to choose
- you need a well-cited baseline for stereo depth estimation research
- you want to train or evaluate stereo matching on KITTI or Scene Flow
- you need a PyTorch reference implementation of PSMNet

## When to avoid
- you need real-time stereo matching on embedded hardware
- you want a maintained production-ready library with recent updates
- you work with monocular (single-camera) depth estimation

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, deep-learning
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: stereo-matching, stereo-vision, depth-estimation, pytorch, disparity, cvpr-2018, research-code, linux, gpu

## Member repositories
- JiaRenChang/PSMNet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:01.826854+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 2026-08-30T04:30:25.017043+00:00, confidence not recorded.
  - readme: https://github.com/JiaRenChang/PSMNet (fetched 2026-08-28T04:05:01.826854+00:00, sha 9bc9d16e44eb)
- Data as of 2026-08-30T08:39:29.467469+00:00.