To retain details while removing artifacts caused by dxt compression on textures

DEDXT

Attempts to remove PVRTC compression artifacts. If 1x_DEPVR is unable to remove some artifacts you should use the other version but it can't remove color artifacts properly so you should overlay color of normal version to it.

DEPVR

DEPVR Artifact

Removing compression artifacts from DXT1 compressed textures. This model was created to remove DXT1 compression artifacts from textures imported into the Source Engine. Compressed textures in the engine sometimes have a green-tint which this model also corrects. The data for this model contained a good mix of diffuse textures and normal maps which means this model is pretty good at removing compression from normals as well. Creating this model was a real learning experience for me and I hope someone finds a good use for it.

DXTDecompressor Source V3

This model is made for Source Engine textures. It tries to remove compression artifacts such as blockyness, discoloration, green tint. It does pretty well on a lot of things, realistic stuff as well, but it was mostly made to work on TF2 textures. It's made to keep as much detail as possible, without any unnecessary denoising/sharpening. Huge thanks to Twittman for assisting me along this journey.

DXTless SourceEngine

Category: Normal Map/Bump Map Generation
Purpose: Map Generation - roughness and displacement maps

This model generates "Franken Maps" (named after Frankenstein), which is a custom material map combination I made. Basically, the Red channel of RGB is just the texture converted to grayscale, the Green channel is the roughness map, and the Blue channel is the displacement map. I had to do this to get around the current limitation of CX loss where it requires a 3 channel output (otherwise I would have just made a 2 channel model, or separate single channel models). As of right now the channels need to be manually split from each other but I will be making a tool for doing this automatically in the coming days.

Pretrained model: 1x_DIV2K-Lite_450k.pth

FrankenMapGenerator-CX Lite

Removes banding, blocking, dithering, aliasing, noise and color tint on DDS Compressed Skin Diffuse Textures.

This should work extremely well on most modern DDS compression types. The training set was compressed with BC3/DXT5, BC3/DXT5 Fast, BC2/DXT3, BC2/DXT3 Fast, and a small number of JPEG compressed images to cover outliers.

This model is trained to remove the slight green color tint that DDS compression tends to add to skin textures, so the model output will not match the original color tone of the input image. This is the desired result though, as DDS compression shifts the colors to a sickly green tint and this model corrects that to more natural color tones.

The training set included faces, body parts, eyes, mouths and hair in a variety of skin types and tones so it should work well on most related diffuse textures.

However it's not just limited to skin, many other images and textures can be cleaned with this model. Designed to be used as a first step cleaning pass before applying additional models after. Check out the other ITF Models.

ITF SkinDiffDDS v1

Category: Skin
Purpose: Skin Upscaling

Adds plausible high frequency detail and removes subtle blur. This is an early unfinished attempt at a x1 Lite model designed specifically for enhancing detail on skin diffuse textures of 3d characters. Even in its current state it works quite well.

Best suited for uncompressed or cleaned textures - otherwise it may just enhance any existing compression artefacts too. The training set included faces, body parts, eyes and hair in a variety of skin types and tones so it should work well on most related diffuse textures.

However it's not just limited to skin, many other images and textures can be enhanced with this model. The results are subtle, so run multiple times if desired.

Pretrained model: 50/50 Interpolation of DIV2K-Lite and SpongeBC1-Lite

ITF SkinDiffDetail Lite v1

N64 textures use a color depth of 5-bits per channel, this model attempts to clean them, restoring smooth gradients in textures

N64clean

Category: Normal Map/Bump Map Generation
Purpose: Map Generation - normal maps

Generating normal maps from textures

Pretrained model: 1x_DIV2K-Lite_450k.pth

NormalMapGenerator-CX Lite

This is part of a larger model set, PBRify_Remix. PBRify_Remix is an easy way to upscale and generate PBR textures using existing low quality textures. The dataset consists of ethically sourced textures from websites like ambientCG and Polyhaven, which sets it apart from most other models.

It's intended for use with RTX Remix (hence the name) but it'll work for other things as well.

Note: The height map model is best used after applying a "delighting" pipeline to your game textures. Currently PBRify_Remix does not have a method for this, so you need your own.

1x-PBRify_Height

This is part of a larger model set, PBRify_Remix. PBRify_Remix is an easy way to upscale and generate PBR textures using existing low quality textures. The dataset consists of ethically sourced textures from websites like ambientCG and Polyhaven, which sets it apart from most other models.

It's intended for use with RTX Remix (hence the name) but it'll work for other things as well.

1x-PBRify_NormalV3

1x-PBRify_RoughnessV2.pth

A 1x model designed for skin contrast (although some backgrounds suffer a little modification), also try the other SkinContrast models

1xSkinContrast-High-SuperUltraCompact

A 1x model designed for skin contrast (although some backgrounds suffer a little modification), some images may suffer from the creation of artifacts, High Alternative is the model that generates the most artifacts, also try the other SkinContrast models

1xSkinContrast-HighAlternative-SuperUltraCompact

1xSkinContrast-SuperUltraCompact

Category: Normal Map/Bump Map Generation
Purpose: Map Generation - Normal Maps

This model generates "Franken Maps" (named after Frankenstein), which is a custom material map combination I made. Basically, the Red channel of RGB is just the texture converted to grayscale, the Green channel is the roughness map, and the Blue channel is the displacement map. I had to do this to get around the current limitation of CX loss where it requires a 3 channel output (otherwise I would have just made a 2 channel model, or separate single channel models). As of right now the channels need to be manually split from each other but I will be making a tool for doing this automatically in the coming days.

Normals Generator General

[Link to Github Release with more infos to the process](https://github.com/Phhofm/models/releases/tag/2xAoMR_mosr)

2xAoMR_mosr  
Scale: 4  
Architecture: [MoSR](https://github.com/umzi2/MoSR)  
Architecture Option: [mosr](https://github.com/umzi2/MoSR/blob/95c5bf73cca014493fe952c2fbc0bdbe593da08f/neosr/archs/mosr_arch.py#L117)  

Author: Philip Hofmann  
License: CC-BY-0.4  
Purpose: A 2x mosr upscaling model for game textures  
Subject: Game Textures  
Input Type: Images  
Release Date: 21.09.2024 (dd/mm/yy)  

Dataset: Game Textures from Age of Mythology: Retold  
Dataset Size: 13'847  
OTF (on the fly augmentations): No  
Pretrained Model: [4xNomos2_hq_mosr](https://github.com/Phhofm/models/releases/tag/4xNomos2_hq_mosr)  
Iterations: 510'000  
Batch Size: 4  
Patch Size: 64  

## Description:   
In short: A 2x game texture mosr upscaling model, trained on and for (but not limited to) Age of Mythology: Retold textures.

Since I have been playing Age of Mythology: Retold (casual player), I thought it would be interesting to train an single image super resolution model on (and for) game textures of AoMR, but this model should be usable for other game textures aswell.
This is a 2x model, since the biggest texture images are already 4096x4096, I thought going 4x on those would be overkill (also there are already 4x game texture upscaling models, so this model can be used for similiar cases where 4x is not needed).

Model Showcase:
[Slowpics](https://slow.pics/s/IlkmsToH)

2xAoMR_mosr

A "lite" model version of my Faithful model

Faithful Lite

Mainly just a test for SPSR. Seems to work better than the original 2xFaithful32_1316 that I used as a pretrained, even though it uses the same dataset.

Pretrained model: 2xFaithful32_1316

FaithfulSPSR

An attempt at recreating the "faithful" style without using the faithful dataset -- aka keeping the "pixel art" style of pixel art.

FakeFaith Lite

Gamecube and Wii textures Dumps.

It is designed as a faster alternative to the full size model, it requires much less Vram and is significantly faster.

HDCube Compact

This is a model for the restoration of My Little Pony: Friendship is Magic, however it also works decently well on all similar art. 

It was trained in 2x on ground truth 3840x2160 HRs and 1920x1080 LRs of varying compression, so it is able to upscale from 1080p to 2160p, where its detail retention is great, however it may create noticeable artifacting if looked at closely, like areas of randomly coloured pixels along edges. In 1x or 1.5x (2x upscaled and then downscaled back down) it performs extremely well, almost perfectly in fact, in correcting colours, removing compression, and crisping up lines - and this is the way the model is intended to be used (hence the acronym of its name being "SS", or "supersampling").

**[Github Release](https://github.com/Derpiesaurus/models/releases/tag/v1.0)**   
**Showcase:** https://slow.pics/s/1ixqCSjy

MLP StarSample V1.0

Fine-tuned 4xFSDedither to upscale images from the game Riven, but should be better in general, particularly on ordered dithering. I adjusted the dataset to have a better variety of dithering parameters, and turned up the HFEN and pixel loss to get better details and color restoration with less noise.

FSDedither Riven

Images from the Fate/Extra games. Works better on the newer ones.

Just a model I trained quickly to test traiNNer-redux and see what I could do with OTF. Works well enough

Architecture	ESRGAN
Scale	1x
Size	32nf12nb
Color Mode	RGB
License	CC-BY-NC-SA-4.0 Private use Distribution Modifications Credit required Same License State Changes No Liability & Warranty Disclaimer
Date	2020-11-04
Dataset	CC0 textures/normals, 2K size, randomly downscaled OTF (by 1, 2, and 4) to get more variation
Dataset size	14848
Training iterations	200000
Training epochs	149
Training HR size	64

NormalMapGenerator-CX Lite

Similar Models