---
license: cc-by-4.0
pipeline_tag: image-to-image
tags:
- pytorch
- super-resolution
---

[Link to Github Release](https://github.com/Phhofm/models/releases/tag/4xBHI_dat2_real)

Scale: 4x  
Network type: DAT2  
Author: Philip Hofmann  
License: CC-BY-4.0  
Purpose: 4x upscaling images. Handles realistic noise, some realistic blur, and webp and jpg (re)compression.  
Pretrained Model: 4xNomos2_hq_dat2  
Training iterations: 240000  
Description: 4x dat2 upscaling model for web and realistic images. It handles realistic noise, some realistic blur, and webp and jpg (re)compression. Trained on my BHI dataset (390'035 training tiles) with degraded LR subset.  

Visual Examples:    
[slow.pics](https://slow.pics/c/WekJvaLo?image-fit=cover)   
[imgsli](https://imgsli.com/MzI3MzI2)   


![Example1](https://github.com/user-attachments/assets/1aab1ed9-dd28-45a6-88db-5317e391bcf4)
![Example2](https://github.com/user-attachments/assets/c5237d90-71a9-4ee9-927c-343ea751d968)
![Example3](https://github.com/user-attachments/assets/34198bbb-66af-46ab-ad9f-56be250b9386)
![Example4](https://github.com/user-attachments/assets/e1a403fa-6035-435a-8f7a-2288047cbf17)
![Example5](https://github.com/user-attachments/assets/8bb0a7fc-300f-4e68-af47-7705fc587533)
![Example6](https://github.com/user-attachments/assets/fd3abfe9-ba2f-4cee-847f-8c2ae19f2c9d)
![Example7](https://github.com/user-attachments/assets/6eb94256-b1bf-4375-8ab5-adfb879849e0)
![Example8](https://github.com/user-attachments/assets/9854e2c7-d009-411c-8f2b-cb1f8c0850d5)
![Example9](https://github.com/user-attachments/assets/cbe09334-6e79-4435-960d-606d7bee5234)
![Example10](https://github.com/user-attachments/assets/08f4d322-dd1f-4275-acd4-bddb8fda9d52)