segmentation_models.pytorch
Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.
- 12k
- Python
- MIT
- 3 d ago
Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.
Pretrained DeepLabv3 and DeepLabv3+ for Pascal VOC & Cityscapes
SSSegmentation: An Open Source Supervised Semantic Segmentation Toolbox Based on PyTorch.
LightNet: Light-weight Networks for Semantic Image Segmentation (Cityscapes and Mapillary Vistas Dataset)
A c++ trainable semantic segmentation library based on libtorch (pytorch c++). Backbone: VGG, ResNet, ResNext. Architecture: FPN, U-Net, PAN, LinkNet, PSPNet, DeepLab-V3, DeepLab-V3+ by now.
📸 PyTorch implementation of MobileNetV3 for real-time semantic segmentation, with pretrained weights & state-of-the-art performance
DeepLabv3+ inference and training in MATLAB for Semantic Segmentation
quick and dirty benchmark for TFLite gles delegate on iOS
Python program to visualize Deeplab (trained on Cityscapes dataset) results.
Learn how to change background efficiently and easily by just using 3-4 lines of code using Python's Pixellib library that uses Deeplab's pre-trained Deep Learning model.
Homography-based perspective video transition using LoFTR feature matching and DeepLabV3 person segmentation for seamless foreground-aware stitching between clips.
Python desktop app example of remove background from image with OpenCV, PyTorch
DeepLab V3 TFLite models that be fully delegated to NNAPI
An AICrowd Challenge: CNN classifier that predicts whether the pixels of an image represent a road or not.
UNet and DeepLabV3 for instance segmentation. Starting from a rgb image of a pet the network should output a segmentation mask of the pet. Deep Learning and Generative Models' practical project developed in 2025.