# PSPNet

Semantic Segmentation in Pytorch

Repository: https://github.com/hszhao/semseg
Canonical: https://ross.abutalabs.com/products/pspnet
Language: Python
License: MIT
License Family: permissive
Last push: 2022-08-28T10:50:55+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": 2918, "days_push": 1466, "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 1378, forks 243 (observed 2026-08-28T04:04:33.526916+00:00)

## What it is
Reference implementation of the Pyramid Scene Parsing Network (PSPNet), a CVPR 2017 semantic segmentation model that won the ImageNet Scene Parsing Challenge 2016. The original repo provides Caffe/Matlab training and evaluation code, with a modernized PyTorch codebase (semseg) also available from the same author.

## Use cases
- train a semantic segmentation model on ADE20K, PASCAL VOC, or Cityscapes
- run scene parsing inference on images with pretrained PSPNet models
- reproduce CVPR 2017 PSPNet research results
- evaluate segmentation models with mIoU metrics
- use PSPNet or PSANet in a PyTorch training pipeline
- benchmark pyramid pooling architectures for dense prediction

## When to choose
- you need the official reference implementation of PSPNet for research reproduction
- you want pretrained models for ADE20K, VOC2012, or Cityscapes segmentation
- you are studying pyramid pooling module architectures

## When to avoid
- you need a production-ready, actively maintained segmentation library
- you cannot build legacy Caffe with CUDA 7-era dependencies
- you want a simple pip-installable tool rather than research code

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: cpp, python
- tags: semantic-segmentation, pspnet, caffe, pytorch, scene-parsing, cvpr2017, research-code, linux, gpu

## Member repositories
- hszhao/semseg (main) score 32
- hszhao/PSPNet (mirror) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:33.526916+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 2026-08-30T03:44:32.822750+00:00, confidence not recorded.
  - readme: https://github.com/hszhao/semseg (fetched 2026-08-28T04:04:33.526916+00:00, sha cd927f3b9083)
- Data as of 2026-08-30T08:39:29.467469+00:00.