# meetps/pytorch-semseg

Semantic Segmentation Architectures Implemented in PyTorch

Repository: https://github.com/meetps/pytorch-semseg
Canonical: https://ross.abutalabs.com/products/pytorch-semseg
Homepage: https://meetshah.dev/semantic-segmentation/deep-learning/pytorch/visdom/2017/06/01/semantic-segmentation-over-the-years.html
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, semantic-segmentation, deep-learning, fully-convolutional-networks
Last push: 2023-10-11T05:27:04+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3451, "days_push": 1057, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3402, forks 787 (observed 2026-08-28T04:08:02.954059+00:00)

## What it is
A PyTorch library implementing popular semantic segmentation architectures such as FCN, U-Net, SegNet, PSPNet, ICNet, FRRN, and LinkNet, with dataloaders for common segmentation datasets. It provides configuration-driven training and validation for pixel-wise dense classification tasks.

## Use cases
- train a semantic segmentation model in pytorch
- implement FCN or U-Net for pixel-wise image classification
- segment images on Pascal VOC or Cityscapes datasets
- compare semantic segmentation architectures like PSPNet and SegNet
- run image segmentation with pretrained encoder backbones
- fine-tune a segmentation network on a custom dataset

## When to choose
- you need reference implementations of classic segmentation architectures in PyTorch
- you want config-driven training on standard segmentation benchmarks like Cityscapes or ADE20K
- you are researching or benchmarking encoder-decoder segmentation models

## When to avoid
- you need state-of-the-art transformer-based segmentation models like SegFormer or Mask2Former
- you require active maintenance and support for recent PyTorch versions
- you want production-ready inference serving rather than research code

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, image-processing, computer-vision
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python, windows
- tags: semantic-segmentation, pytorch, fcn, unet, pspnet, segnet, linknet, icnet, frrn, computer-vision, linux, macos, gpu

## Member repositories
- meetps/pytorch-semseg (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:02.954059+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 2026-08-29T18:38:38.765390+00:00, confidence not recorded.
  - readme: https://github.com/meetps/pytorch-semseg (fetched 2026-08-28T04:08:02.954059+00:00, sha 6d28b1cf7c23)
  - homepage: https://meetshah.dev/semantic-segmentation/deep-learning/pytorch/visdom/2017/06/01/semantic-segmentation-over-the-years.html (fetched 2026-08-29T09:32:54.660813+00:00, sha a59aba628417)
- Data as of 2026-08-30T08:39:29.467469+00:00.