SwinIR
Image restoration using Swin Transformer for super-resolution and denoising.
About
SwinIR applies the Swin Transformer with shifted-window attention to image restoration tasks including super-resolution, denoising, and JPEG artifact removal. It reaches strong results across standard benchmarks and ships pretrained models and a web demo, with related VRT and RVRT projects extending the approach to video. Developed at the Computer Vision Lab, ETH Zurich. Released under the Apache 2.0 license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- AI Image/Video Editing
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- Apache-2.0
- Minimum VRAM
- 4 GB
- Added
- Apr 3, 2026
Related Tools
Practical image super-resolution model trained on diverse degradations.
Robust face restoration using codebook lookup transformer.
One-click Stable Diffusion app for macOS with Apple Silicon optimization.
AI-powered image inpainting tool for removing objects and defects from photos.
Optimized fork of AUTOMATIC1111 with better memory management and speed.
Feature-rich web UI for Stable Diffusion with extensive extension ecosystem.