[Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (ICCV, 2021)](https://github.com/cszn/BSRGAN)

BSRGANx2

SPAN model for live action film and digital video. The main goal is to fix/reduce common video quality problems while maintaining fidelity. I tried the existing video-focused models and they all denoise or cause colour shifts so I decided to train my own.

The model is trained with compression (JPEG, MPEG-4 ASP, H264, VP9, H265), chroma subsampling, blurriness from multiple scaling, uneven horizontal and vertical resolution, oversharpening halos, bad deinterlacing jaggies, and onscreen text. It is not trained to remove noise at all so it preserves details in the source well. To prevent colour/brightness shifts, I used consistency loss in neosr. I had to modify consistency loss to use a stronger blur so it doesn't interfere with the halo removal.

Limitations:
1. The model has limited ability to see details through heavy grain, but light to moderate grain is fine.
2. The model still does not handle bad deinterlacing perfectly, especially if the source is vertically resized. Fixing bad deinterlacing is not the main goal so it is what it is. Sources that are line-doubled throughout should be descaled back to half height first for best results.
3. The model sometimes oversharpens a little. This is probably because the training data has some oversharpened images.
4. This model generally cannot handle VHS degradation.

More comparisons: https://slow.pics/c/DtDN7gaq

The training config and image degradation scripts used to create training data can be found in https://github.com/jcj83429/upscaling/tree/9332e7d5b07747ff347e5abdc43f8144364de9f7/2xLiveActionV1_SPAN

2xLiveActionV1_SPAN

[Omni Aggregation Networks for Lightweight Image Super-Resolution (OmniSR)](https://github.com/Francis0625/Omni-SR)

While lightweight ViT framework has made tremendous progress in image super-resolution, its uni-dimensional
self-attention modeling, as well as homogeneous aggregation scheme, limit its effective receptive field (ERF) to
include more comprehensive interactions from both spatial and channel dimensions. To tackle these drawbacks,
this work proposes two enhanced components under a new Omni-SR architecture. First, an Omni Self-Attention (OSA) block is proposed based on dense interaction principle, which can simultaneously model pixel-interaction from both spatial and channel dimensions, mining the potential correlations across omni-axis (i.e., spatial and channel). Coupling with mainstream window partitioning strategies, OSA can achieve superior performance with compelling computational budgets. Second, a multi-scale interaction scheme is proposed to mitigate sub-optimal ERF (i.e., premature saturation) in shallow models, which facilitates local propagation and meso-/global-scale interactions, rendering an omni-scale aggregation building block. Extensive experiments demonstrate that Omni-SR achieves recordhigh performance on lightweight super-resolution benchmarks (e.g., 26.95dB@Urban100 ×4 with only 792K parameters). Our code is available at https://github.com/Francis0625/Omni-SR

OmniSR 2x DF2K

OmniSR 2x DIV2K

A 2x photo upscaling compact model based on microsofts image pairs. This was one of the earliest models I started training and finished it now for release.

Parimg Compact

Purpose: VHS Restoration

An advanced VHS recording model designed to enhance video quality by reducing artifacts such as haloing, ghosting, and noise patterns. Optimized primarily for PAL resolution (NTSC might work good as well).

VHS2HD

[SwinIR: Image Restoration Using Swin Transformer](https://github.com/JingyunLiang/SwinIR)


Image restoration is a long-standing low-level vision problem that aims to restore high-quality images from low-quality images (e.g., downscaled, noisy and compressed images). While state-of-the-art image restoration methods are based on convolutional neural networks, few attempts have been made with Transformers which show impressive performance on high-level vision tasks. In this paper, we propose a strong baseline model SwinIR for image restoration based on the Swin Transformer. SwinIR consists of three parts: shallow feature extraction, deep feature extraction and high-quality image reconstruction. In particular, the deep feature extraction module is composed of several residual Swin Transformer blocks (RSTB), each of which has several Swin Transformer layers together with a residual connection. We conduct experiments on three representative tasks: image super-resolution (including classical, lightweight and real-world image super-resolution), image denoising (including grayscale and color image denoising) and JPEG compression artifact reduction. Experimental results demonstrate that SwinIR outperforms state-of-the-art methods on different tasks by up to 0.14~0.45dB, while the total number of parameters can be reduced by up to 67%.

SwinIR-M-x2 (classicalSR-DF2K-s64w8)

SwinIR-M-x2 (classicalSR-DIV2K-s64w8)

SwinIR-M-x2 (lightweightSR-DIV2K-s64w8)

SwinIR-M-x2-GAN (realSR-BSRGAN-DFO-s64w8)

[Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) aims at developing Practical Algorithms for General Image/Video Restoration.

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
We extend the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data.

RealESRGAN_x2Plus

[Real-ESRGAN](https://github.com/xinntao/Real-ESRGAN) aims at developing Practical Algorithms for General Image/Video Restoration.

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
We extend the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data.

We add small models that are optimized for anime videos :-)

RealESRGANv2 AnimeVideo xs-x2

Description: 4x photo upscaler for faces, trained on the FaceUp dataset. These models are an improvement over the previously released 4xFFHQDAT and are its successors. These models are released together with the FaceUp dataset, plus the accompanying [youtube video](https://www.youtube.com/watch?v=TBiVIzQkptI)

This model comes in 4 different versions:  
4xFaceUpDAT (for good quality input)  
4xFaceUpLDAT (for lower quality input, can additionally denoise)  
4xFaceUpSharpDAT (for good quality input, produces sharper output, trained without USM but sharpened input images, good quality input)  
4xFaceUpSharpLDAT (for lower quality input, produces sharper output, trained without USM but sharpened input images, can additionally denoise)  

I recommend trying out 4xFaceUpDAT

4xFaceUpDAT

4xFaceUpSharpDAT

A normal ESRGAN model without degradations and without any pretrain, simply an RRDBNet model trained on paired dataset (4x downsampled) on the full LSDIR dataset (84,991 images / 165 GB)

4xLSDIR

Upscale small good quality photos to 4x their size

My first ever model 😄 Well, it’s not the best, but, it’s something 😉 
I provide some 15 examples from the validation set here for you to visually see the generated output (with chaiNNer), photo dimensions are in the name

LSDIR Compact

Upscale photos to x4 their size.

4xLSDIRCompactv2 supersedes the previously released models, it combines all my progress on my compact model. Both CompactC and CompactR had received around 8 hours more training since release with batch size 10 (CompactR had only been up to 5 on release), and these two were then interpolated together. This allows v2 to handle some degradations, while preserving the details of the CompactC model. Examples: https://imgsli.com/MTY0Njgz/0/2

LSDIR Compact v2

Upscale small photos with compression to 4x their size.

Trying to extend my previous model to be able to handle compression (JPG 100-30) by manually altering the training dataset, since 4xLSDIRCompact can't handle compression. Use this instead of 4xLSDIRCompact if your photo has compression (like an image from the web).

LSDIR Compact C

Upscale compressed photos to x4 their size.

Able to handle JPG compression (30-100).

LSDIR Compact C3

Upscale good quality input photos to x4 their size

The original 4xLSDIRCompact a bit more trained, cannot handle degradation

LSDIRCompactN

Upscale small photos with compression, noise and slight blur to 4x their size.

Extending my last 4xLSDIRCompact model to Real-ESRGAN, meaning trained on synthetic data instead to handle more kinds of degradations, it should be able to handle compression, noise, and slight blur.

Here is a comparison to show that 4xLSDIRCompact cannot handle compression artifacts, and that these two models will produce better output for that specific scenario. These models are not ‘better’ than the previous one, they are just meant to handle a different use case.

LSDIR Compact R

Upscale (degraded) photos to x4 their size. 

Trained on synthetic data, meant to handle more degradations

LSDIR Compact R3

The RealESRGAN_x4plus finetuned with the big LSDIR dataset (84,991 images / 165 GB), with manually added jpg compression.

4xLSDIRplusC

The RealESRGAN_x4plus finetuned with the big LSDIR dataset (84,991 images / 165 GB), no degradation.

4xLSDIRplusN

[Link to Github Release](https://github.com/Phhofm/models/releases/tag/4xDRCT-mssim-pretrains)

Since no [DRCT](https://github.com/ming053l/DRCT/tree/main) model releases exist yet, I am releasing my drct-s, drct and drct-l pretrains. 
These can be used to speed up and stabilize early training stages when training new drct models.
Trained with mssim (and augs, and color and lumaloss) on downscaled [nomosuni](https://github.com/muslll/neosr/tree/0ecbeb98dcefd97801da0ea39ca2cd03a87fce21?tab=readme-ov-file#datasets).
Training files are also provided additionally.

## 4xmssim_drct-l_pretrain

**Scale:** 4  
**Architecture:** [DRCT](https://github.com/ming053l/DRCT)  
**Architecture Option:** [DRCT-L](https://github.com/ming053l/DRCT/blob/8c13e63135f494913cd504c073e41ef52250d1d4/options/train/train_DRCT-L_SRx4_ImageNet_from_scratch.yml#L55)  

**Author:** Philip Hofmann  
**License:** CC-BY-0.4  
**Purpose:** Pretrained  
**Subject:** Anime, Photography  
**Input Type:** Images  
**Release Date:** 28.04.2024  

**Dataset:** [nomos_uni](https://github.com/muslll/neosr?tab=readme-ov-file#datasets)  
**Dataset Size:** 2989  
**OTF (on the fly augmentations):** No  
**Pretrained Model:** None (=From Scratch)  
**Iterations:** 108'000  
**Batch Size:** 2-6  
**GT Size:** 128-320  

**Description:** A pretrain to start drct-l model training, since there are no official released drct pretrains yet I trained these myself and release them here.

Architecture	Compact
Scale	2x
Size	64nf16nc
Color Mode	RGB
License	CC-BY-4.0 Private use Commercial use Distribution Modifications Credit required State Changes No Liability & Warranty Disclaimer
Date	2023-05-05
Dataset	Microsoft ImagePairs
Training iterations	100000
Training HR size	256
Training OTF	No

Parimg Compact

Similar Models