This repository provides semantic segmentation models with more than 500 pretrained convolutional and transformer backbones.
Open source
24 alternativessegmentation models
This repository provides segmentation models with pretrained backbones for Keras and TensorFlow Keras.
About segmentation models
segmentation_models is an open source Python library for image segmentation built for Keras and TensorFlow Keras. It provides 4 standard segmentation model architectures (Unet, FPN, Linknet, PSPNet) with 25 pretrained backbones per architecture, plus built-in segmentation losses (Jaccard, Dice, Focal) and metrics (IoU, F-score). Pretrained encoder weights enable faster model convergence, though version 1. models are not backward compatible with models trained on version 0. releases, which can be accessed by installing version 0.2.1.
Key features
- High-level API for building segmentation models with minimal code
- 4 segmentation model architectures (Unet, FPN, Linknet, PSPNet)
- 25 pretrained backbones available per model architecture
- Pretrained encoder weights for faster model convergence
- Built-in segmentation losses and metrics (Jaccard, Dice, Focal, IoU, F-score)
- Supports both Keras and TensorFlow Keras frameworks
Best for: Developers and researchers building image segmentation models with Keras or TensorFlow Keras.
Good for
- Open source and freely available for use and modification
- Pretrained backbones reduce training time and improve convergence
- Supports multiple common segmentation architectures and Keras frameworks
Watch out for
- Version 1.* models are not backward compatible with 0.* trained models
- Pretrained encoder weights cannot be used for non-3-channel input shapes
segmentation models at a glance
- Category
- AI & Machine Learning › AI Data & Analytics
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
segmentation models alternatives
Tools and sites you could use instead, ranked by how closely they match.
Spektral provides a Python framework for creating graph neural networks with Keras and TensorFlow.
Open source
24 alternatives
Ultralytics supports detection, segmentation, image classification, pose estimation, and tracking across YOLO model versions.
Open source
24 alternatives
Ultralytics YOLOv5 supports object detection, instance segmentation, classification, training, and export in PyTorch.
Open source
24 alternatives
Detectron2 provides computer vision tools for object detection, segmentation, and other visual recognition tasks.
Open source
24 alternatives
This repository implements YOLOv3 variants in PyTorch for real time object detection, training, validation, inference, and export.
Open source
24 alternativesPopular in AI Data & Analytics
Well-known listings from the same topic.
Dotgetdot.aiForvioforvio.comFluidAI Medicalfluidai.mdApache Zeppelinzeppelin.apache.orgSola Securitysola.securityQuadraticquadratichq.com
segmentation models FAQ
Which deep learning frameworks does the library support?
It works with Keras and TensorFlow Keras, with options to switch between them via environment variable or set_framework method.
What segmentation model architectures are included?
The library includes Unet, FPN, Linknet and PSPNet architectures.
Can pretrained weights be used for custom input channel counts?
No, pretrained encoder weights only work for 3-channel inputs; set encoder_weights=None for other channel counts.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.