mv-lab/swin2sr
A SwinV2 transformer model for reconstructing high-quality images from compressed or low-resolution inputs.
Not currently ranked — collecting fresh signals.
star history
Swin2SR is a computer vision model that performs image super-resolution and restoration on compressed images using a Swin transformer architecture. The model was developed for the AIM workshop at ECCV 2022 and supports tasks including JPEG artifact reduction, denoising, and general image enhancement. It provides pretrained weights and inference code for applying the model to real-world image restoration tasks.
Frequently asked
- What is mv-lab/swin2sr?
- A SwinV2 transformer model for reconstructing high-quality images from compressed or low-resolution inputs.
- Is swin2sr open source?
- Yes — mv-lab/swin2sr is open source, released under the Apache-2.0 license.
- What language is swin2sr written in?
- mv-lab/swin2sr is primarily written in Python.
- How popular is swin2sr?
- mv-lab/swin2sr has 689 stars on GitHub.
- Where can I find swin2sr?
- mv-lab/swin2sr is on GitHub at https://github.com/mv-lab/swin2sr.