# qubvel/segmentation_models

Segmentation models with pretrained backbones. Keras and TensorFlow Keras.

Repository: https://github.com/qubvel/segmentation_models
Canonical: https://ross.abutalabs.com/products/segmentation_models
Language: Python
License: MIT
License Family: permissive
Topics: unet, fpn, segmentation, keras, pretrained, pre-trained, image-segmentation, linknet, pspnet, tensorflow, segmentation-models, resnet, resnext, efficientnet, densenet, keras-tensorflow, keras-models, tensorflow-keras, keras-examples, mobilenet
Last push: 2024-08-21T11:16:16+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3011, "days_push": 742, "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 4923, forks 1041 (observed 2026-08-28T04:09:02.481982+00:00)

## What it is
A Python library providing neural network architectures for image segmentation (Unet, FPN, Linknet, PSPNet) built on Keras and TensorFlow Keras. It offers 25 pretrained backbones per architecture plus segmentation losses and metrics, with a two-line high-level API.

## Use cases
- train an image segmentation model with pretrained backbones
- build a UNet in Keras with two lines of code
- semantic segmentation of medical images
- binary and multi-class pixel-wise classification
- use EfficientNet or ResNet encoders for segmentation
- apply Dice and Focal losses for segmentation training

## When to choose
- you need ready-made segmentation architectures with pretrained encoders in Keras/TensorFlow
- you want a simple high-level API for image segmentation experiments
- you need segmentation-specific losses and metrics out of the box

## When to avoid
- you work in PyTorch (use segmentation_models.pytorch instead)
- you need cutting-edge transformer-based segmentation models
- you require frequent updates or new architecture support

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing
- domain: computer-vision, image-processing, deep-learning, machine-learning
- platform: python
- tags: image-segmentation, unet, fpn, pspnet, linknet, pretrained-backbones, keras, tensorflow

## Member repositories
- qubvel/segmentation_models (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:02.481982+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 2026-08-29T18:18:01.118859+00:00, confidence not recorded.
  - readme: https://github.com/qubvel/segmentation_models (fetched 2026-08-28T04:09:02.481982+00:00, sha 3ad08333d6c4)
  - registry_pypi: https://pypi.org/pypi/segmentation_models/json (fetched 2026-08-29T08:59:48.477311+00:00, sha f24c1cab1fe9)
- Data as of 2026-08-30T08:39:29.467469+00:00.