ZhendongWang6/Uformer
A U-shaped Transformer architecture for image restoration tasks including deblurring, denoising, deraining, and demoireing.
Not currently ranked — collecting fresh signals.
star history
Uformer is a Transformer-based neural network designed for various image restoration tasks. It employs a U-shaped encoder-decoder architecture with self-attention mechanisms to handle tasks like image deblurring, denoising, deraining, and demoireing. The model is implemented in PyTorch and achieves state-of-the-art results on benchmarks like GoPro, RealBlur, DND, and SIDD.
Frequently asked
- What is ZhendongWang6/Uformer?
- A U-shaped Transformer architecture for image restoration tasks including deblurring, denoising, deraining, and demoireing.
- Is Uformer open source?
- Yes — ZhendongWang6/Uformer is open source, released under the MIT license.
- What language is Uformer written in?
- ZhendongWang6/Uformer is primarily written in Python.
- How popular is Uformer?
- ZhendongWang6/Uformer has 945 stars on GitHub.
- Where can I find Uformer?
- ZhendongWang6/Uformer is on GitHub at https://github.com/ZhendongWang6/Uformer.