HuCaoFighting/Swin-Unet
A pure transformer-based U-Net architecture for medical image segmentation using Swin Transformer.
Not currently ranked — collecting fresh signals.
star history
This repository implements Swin-Unet, a vision transformer architecture adapted for medical image segmentation tasks. It uses a Swin Transformer encoder combined with a U-Net style decoder to perform semantic segmentation on medical imaging datasets such as Synapse and ACDC. The model is trained using PyTorch with configurable hyperparameters including batch size, learning rate, and image size.
Frequently asked
- What is HuCaoFighting/Swin-Unet?
- A pure transformer-based U-Net architecture for medical image segmentation using Swin Transformer.
- Is Swin-Unet open source?
- Yes — HuCaoFighting/Swin-Unet is an open-source project tracked on heatdrop.
- What language is Swin-Unet written in?
- HuCaoFighting/Swin-Unet is primarily written in Python.
- How popular is Swin-Unet?
- HuCaoFighting/Swin-Unet has 2.4k stars on GitHub.
- Where can I find Swin-Unet?
- HuCaoFighting/Swin-Unet is on GitHub at https://github.com/HuCaoFighting/Swin-Unet.