YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
SwinIR Low-Light 4x Super-Resolution
This repository contains a fine-tuned SwinIR-M model for low-light image restoration and 4x super-resolution.
Model
Architecture: SwinIR-M Upscale factor: 4 Input channels: 3 Image size: 64 Window size: 8 Embedding dimension: 180 Depths: [6, 6, 6, 6, 6, 6] Number of heads: [6, 6, 6, 6, 6, 6] MLP ratio: 2 Upsampler: PixelShuffle Residual connection: 1conv
Files
- model.pth - trained model weights
- network_swinir.py - model architecture
- config.json - architecture configuration
Training
The model was fine-tuned for low-light image restoration and 4x super-resolution.
Best validation PSNR: 39.1407 dB
- Downloads last month
- 10
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support