|
|
--- |
|
|
license: cc-by-nc-sa-4.0 |
|
|
pipeline_tag: image-to-image |
|
|
tags: |
|
|
- pytorch |
|
|
- super-resolution |
|
|
--- |
|
|
|
|
|
**Name:** 4x-AnimeSharp (aka 4x-TextSharpV1) |
|
|
|
|
|
**License:** CC BY-NC-SA 4.0 |
|
|
|
|
|
**Link:** <https://mega.nz/file/CFZGHb4A#yo2NaVANd9-gk9n4lxA9zkzKPqVukZYpzendRVCinDw> |
|
|
|
|
|
**Model Architecture:** ESRGAN |
|
|
|
|
|
**Scale:** 4 |
|
|
|
|
|
**Purpose:** Anime or Text |
|
|
|
|
|
|
|
|
**Iterations:** Interpolation between 150k and 8k |
|
|
|
|
|
**batch_size:** 4/8 |
|
|
|
|
|
**HR_size:** 128 |
|
|
|
|
|
**Epoch:** Not Relevant |
|
|
|
|
|
**Dataset:** Custom |
|
|
|
|
|
**OTF Training** Yes |
|
|
|
|
|
**Pretrained_Model_G:** Interpolation between 4x-UltraSharp and 4x-TextSharp-v0.5 |
|
|
|
|
|
|
|
|
**Description:** Works amazingly on anime. It also upscales text, but it's far better with anime content than text. |
|
|
|
|
|
I rebranded this model on 2/10/22 to 4x-AnimeSharp from 4x-TextSharpV1. |
|
|
|
|
|
https://cdn.discordapp.com/attachments/903415274521374750/925533775616696340/unknown.png |
|
|
https://cdn.discordapp.com/attachments/549525506585001985/941432207917084743/Clipboard-comparison.png |
|
|
https://cdn.discordapp.com/attachments/549525506585001985/941432638206541825/test_pic-comparison.png |
