This repository provides TensorFlow/Keras implementations of classic and modern segmentation architectures including U-net, V-net, U-net++, UNET 3+, Attention U-net, R2U-Net, ResUnet-a, U^2-Net, TransUNET, and Swin-UNET. These are encoder-decoder convolutional neural networks primarily used for semantic segmentation tasks in images. The library supports optional ImageNet-pretrained backbones and deep supervision mechanisms for several models.