|
--- |
|
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/4xNomosWebPhoto_RealPLKSR) |
|
|
|
# 4xNomosWebPhoto_RealPLKSR |
|
|
|
Scale: 4 |
|
Architecture: [RealPLKSR](https://github.com/muslll/neosr/blob/c8720232448eb059567ae820e3d461d66a4aef1c/neosr/archs/realplksr_arch.py) |
|
Architecture Option: realplksr |
|
|
|
Author: Philip Hofmann |
|
License: CC-BY-0.4 |
|
Purpose: Restoration |
|
Subject: Photography |
|
Input Type: Images |
|
Release Date: 28.05.2024 |
|
|
|
Dataset: [Nomos-v2](https://github.com/muslll/neosr?tab=readme-ov-file#-datasets) |
|
Dataset Size: 6000 |
|
OTF (on the fly augmentations): No |
|
Pretrained Model: 4x_realplksr_gan_pretrain |
|
Iterations: 404'000, 445'000 |
|
Batch Size: 12, 4 |
|
GT Size: 128, 256, 512 |
|
|
|
Description: |
|
|
|
short: 4x RealPLKSR model for photography, trained with realistic noise, lens blur, jpg and webp re-compression. |
|
|
|
full: My newest version of my RealWebPhoto series, this time I used the newly released [Nomos-v2](https://github.com/muslll/neosr?tab=readme-ov-file#-datasets) dataset by musl. |
|
I then made 12 different low resolution degraded folders, using [kim's datasetdestroyer](https://github.com/Kim2091/helpful-scripts/tree/main/Dataset%20Destroyer) for scaling and compression, my [ludvae200 model](https://github.com/Phhofm/models/releases/tag/Ludvae200) for realistic noise, and [umzi's wtp_dataset_destroyer](https://github.com/umzi2/wtp_dataset_destroyer/tree/master?tab=readme-ov-file) with its floating point lens blur implementation for better control (since i needed to control the lens blur strength more precisely). |
|
I then mixed them together in a single lr folder and trained for 460'000 iters, checked the results, and upon kims suggestion of using interpolation, I tested and am releasing this interpolation between the checkpoints 404'000 and 445'000. |
|
|
|
This model has been trained on [neosr](https://github.com/muslll/neosr) using mixup, cutmix, resizemix, cutblur, nadam, unet, multisteplr, mssim, perceptual, gan, dists, ldl, ff, color and lumaloss, and interpolated using the current [chaiNNer](https://github.com/chaiNNer-org/chaiNNer) nightly version. |
|
|
|
This model took multiple retrainings and reworks of the dataset, until I am now satisfied enough with the quality to release this version. |
|
|
|
For more details on the whole process see the [pdf file](https://huggingface.co/Phips/4xNomosWebPhoto_RealPLKSR/blob/main/4xNomosWebPhoto-4.pdf) in the attachement. |
|
|
|
I am also attaching the 404'000, 445'000 and 460'000 checkpoints for completeness. |
|
|
|
PS in general degradation strengths have been reduced/adjusted in comparison to my previous RealWebPhoto models |
|
|
|
Showcase: |
|
[Slow Pics 10 Examples](https://slow.pics/s/euvEv4hL) |
|
|
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
 |
