NCNN implementation of Real-ESRGAN. Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration.

Related tags

Deep Learninganimeamdgpuvulkanintelnvidiasuper-resolutionimage-restorationdenoisencnnesrganreal-esrganjpeg-compresion
Overview

Real-ESRGAN ncnn Vulkan

CI License: MIT Open issue Closed issue

This project is the ncnn implementation of Real-ESRGAN. Real-ESRGAN ncnn Vulkan heavily borrows from realsr-ncnn-vulkan. Many thanks to nihui, ncnn and realsr-ncnn-vulkan 😁

Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration. We also optimize it for anime images.

Contents

If Real-ESRGAN is helpful in your photos/projects, please help to ⭐ this repo or recommend it to your friends. Thanks 😊
Other recommended projects:
▶️ Real-ESRGAN: A practical algorithm for general image restoration
▶️ GFPGAN: A practical algorithm for real-world face restoration
▶️ BasicSR: An open-source image and video restoration toolbox
▶️ facexlib: A collection that provides useful face-relation functions.
▶️ HandyView: A PyQt5-based image viewer that is handy for view and comparison.

📖 Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

[Paper]   [Project Page]   [Demo]
Xintao Wang, Liangbin Xie, Chao Dong, Ying Shan
Tencent ARC Lab; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences

⏳ TODO List

  • Support further cheap arbitrary resize (e.g., bicubic, bilinear) for the model outputs
  • Bug: Some PCs will output black images
  • Add the guidance for ncnn model conversion
  • Support face restoration - GFPGAN

💻 Usages

Example Command

realesrgan-ncnn-vulkan.exe -i input.jpg -o output.png -n realesrgan-x4plus-anime

Full Usages

Usage: realesrgan-ncnn-vulkan.exe -i infile -o outfile [options]...

  -h                   show this help
  -v                   verbose output
  -i input-path        input image path (jpg/png/webp) or directory
  -o output-path       output image path (jpg/png/webp) or directory
  -s scale             upscale ratio (4, default=4)
  -t tile-size         tile size (>=32/0=auto, default=0) can be 0,0,0 for multi-gpu
  -m model-path        folder path to pre-trained models(default=models)
  -n model-name        model name (default=realesrgan-x4plus, can be realesrgan-x4plus | realesrgan-x4plus-anime | realesrnet-x4plus)
  -g gpu-id            gpu device to use (default=0) can be 0,1,2 for multi-gpu
  -j load:proc:save    thread count for load/proc/save (default=1:2:2) can be 1:2,2,2:2 for multi-gpu
  -x                   enable tta mode
  -f format            output image format (jpg/png/webp, default=ext/png)
  • input-path and output-path accept either file path or directory path
  • scale = scale level, 4 = upscale 4x
  • tile-size = tile size, use smaller value to reduce GPU memory usage, default selects automatically
  • load:proc:save = thread count for the three stages (image decoding + model upscaling + image encoding), using larger values may increase GPU usage and consume more GPU memory. You can tune this configuration with "4:4:4" for many small-size images, and "2:2:2" for large-size images. The default setting usually works fine for most situations. If you find that your GPU is hungry, try increasing thread count to achieve faster processing.
  • format = the format of the image to be output, png is better supported, however webp generally yields smaller file sizes, both are losslessly encoded

If you encounter crash or error, try to upgrade your GPU driver

🌏 Other Open-Source Code Used

📜 BibTeX

@InProceedings{wang2021realesrgan,
    author    = {Xintao Wang and Liangbin Xie and Chao Dong and Ying Shan},
    title     = {Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data},
    booktitle = {International Conference on Computer Vision Workshops (ICCVW)},
    date      = {2021}
}

📧 Contact

If you have any question, please email [email protected] or [email protected].

Comments
  • problem running on aws

    problem running on aws

    I downloaded the ubuntu zip on a g3s.xlarge and the result is a black image. Is the zip missing files?

    
./realesrgan-ncnn-vulkan-v0.2.0-ubuntu/realesrgan-ncnn-vulkan -i input.jpg -o out/output.jpg -n realesrgan-x4plus -s 4 

[0 Tesla M60]  queueC=0[16]  queueG=0[16]  queueT=1[2]
[0 Tesla M60]  bugsbn1=0  bugbilz=0  bugcopc=0  bugihfa=0
[0 Tesla M60]  fp16-p/s/a=1/1/0  int8-p/s/a=1/1/1
[0 Tesla M60]  subgroup=32  basic=1  vote=1  ballot=1  shuffle=1
[1 llvmpipe (LLVM 12.0.0, 256 bits)]  queueC=0[1]  queueG=0[1]  queueT=0[1]
[1 llvmpipe (LLVM 12.0.0, 256 bits)]  bugsbn1=0  bugbilz=0  bugcopc=0  bugihfa=0
[1 llvmpipe (LLVM 12.0.0, 256 bits)]  fp16-p/s/a=1/1/0  int8-p/s/a=1/1/0
[1 llvmpipe (LLVM 12.0.0, 256 bits)]  subgroup=8  basic=1  vote=1  ballot=1  shuffle=0
fopen /home/ubuntu/realesrgan-ncnn-vulkan-v0.2.0-ubuntu/models/realesrgan-x4plus.param failed
fopen /home/ubuntu/realesrgan-ncnn-vulkan-v0.2.0-ubuntu/models/realesrgan-x4plus.bin failed
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
0.00%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
4.17%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
8.33%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
12.50%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
16.67%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
20.83%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
25.00%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
29.17%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
33.33%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
37.50%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
41.67%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
45.83%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
50.00%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
54.17%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
58.33%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
62.50%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
66.67%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
70.83%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
75.00%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
79.17%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
83.33%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
87.50%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
91.67%
find_blob_index_by_name data failed
Try
find_blob_index_by_name output failed
Try
95.83%
    opened by kmulvey 2
  • How do I switch from integrated gpu to nvedia gpu

    How do I switch from integrated gpu to nvedia gpu

    [0 Intel(R) UHD Graphics 630] queueC=0[1] queueG=0[1] queueT=0[1] [0 Intel(R) UHD Graphics 630] bugsbn1=0 bugbilz=3 bugcopc=0 bugihfa=0 [0 Intel(R) UHD Graphics 630] fp16-p/s/a=1/1/1 int8-p/s/a=1/1/1 [0 Intel(R) UHD Graphics 630] subgroup=32 basic=1 vote=1 ballot=1 shuffle=1

    When upscaling images it only uses integrated gpu I want to use my gtx 1650 how do I switch this?

    opened by csAshish 0
  • in directory mode, option to skip if destination already exist

    in directory mode, option to skip if destination already exist

    if you stop the upscaler while processing a directory and you have to restart it, it will start from the beginning and overwrite existing files

    overwriting in file mode is fine but should be an option for direcotry mode ... so I propose a flag to disable overwrite and skip if exist

    opened by 6543 0
  • models: consider adding into README how to get them.

    models: consider adding into README how to get them.

    Hey,

    Please consider adding to README a section how to get the pre-trained model files. Currently you'd need to extract them from the following links

    • https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesrgan-ncnn-vulkan-20220424-ubuntu.zip
    • https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth
    • https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth

    It took me a moment to figure out how to get a hold of them and I suspect many others would face similar challenge.

    opened by slashbeast 0
  • Segfault realesrnet-x4plus

    Segfault realesrnet-x4plus 

    %  ./realesrgan-ncnn-vulkan -i FdLbqKcWIAAaCeY.jpeg  -o FdLbqKcWIAAaCeY.png -n realesrnet-x4plus
zsh: segmentation fault  ./realesrgan-ncnn-vulkan -i FdLbqKcWIAAaCeY.jpeg -o FdLbqKcWIAAaCeY.png -n
    opened by atomical 1
Owner
Xintao
Researcher at Tencent ARC Lab, (Applied Research Center)
Xintao
GitHub Repository https://github.com/xinntao/Real-ESRGAN
Fuzzing the Kernel Using Unicornafl and AFL++

Unicorefuzz Fuzzing the Kernel using UnicornAFL and AFL++. For details, skim through the WOOT paper or watch this talk at CCCamp19. Is it any good? ye

Security in Telecommunications 283 Dec 26, 2022
Code for paper "Which Training Methods for GANs do actually Converge? (ICML 2018)"

GAN stability This repository contains the experiments in the supplementary material for the paper Which Training Methods for GANs do actually Converg

Lars Mescheder 885 Jan 01, 2023
CVPR 2021 Official Pytorch Code for UC2: Universal Cross-lingual Cross-modal Vision-and-Language Pre-training

UC2 UC2: Universal Cross-lingual Cross-modal Vision-and-Language Pre-training Mingyang Zhou, Luowei Zhou, Shuohang Wang, Yu Cheng, Linjie Li, Zhou Yu,

Mingyang Zhou 28 Dec 30, 2022
Reference implementation of code generation projects from Facebook AI Research. General toolkit to apply machine learning to code, from dataset creation to model training and evaluation. Comes with pretrained models.

This repository is a toolkit to do machine learning for programming languages. It implements tokenization, dataset preprocessing, model training and m

Facebook Research 408 Jan 01, 2023
ICSS - Interactive Continual Semantic Segmentation

Presentation This repository contains the code of our paper: Weakly-supervised c

Alteia 9 Jul 23, 2022
Advbox is a toolbox to generate adversarial examples that fool neural networks in PaddlePaddle、PyTorch、Caffe2、MxNet、Keras、TensorFlow and Advbox can benchmark the robustness of machine learning models.

Advbox is a toolbox to generate adversarial examples that fool neural networks in PaddlePaddle、PyTorch、Caffe2、MxNet、Keras、TensorFlow and Advbox can benchmark the robustness of machine learning models

AdvBox 1.3k Dec 25, 2022
Unified Pre-training for Self-Supervised Learning and Supervised Learning for ASR

UniSpeech The family of UniSpeech: UniSpeech (ICML 2021): Unified Pre-training for Self-Supervised Learning and Supervised Learning for ASR UniSpeech-

Microsoft 282 Jan 09, 2023
TensorFlow implementation of ENet, trained on the Cityscapes dataset.

segmentation TensorFlow implementation of ENet (https://arxiv.org/pdf/1606.02147.pdf) based on the official Torch implementation (https://github.com/e

Fredrik Gustafsson 248 Dec 16, 2022
Prior-Guided Multi-View 3D Head Reconstruction

Prior-Guided Head MVS This repository includes some reconstruction results of our IEEE TMM 2021 paper, Prior-Guided Multi-View 3D Head Reconstruction.

11 Aug 17, 2022
PyTorch code for 'Efficient Single Image Super-Resolution Using Dual Path Connections with Multiple Scale Learning'

Efficient Single Image Super-Resolution Using Dual Path Connections with Multiple Scale Learning This repository is for EMSRDPN introduced in the foll

7 Feb 10, 2022
Official implementation of the paper DeFlow: Learning Complex Image Degradations from Unpaired Data with Conditional Flows

DeFlow: Learning Complex Image Degradations from Unpaired Data with Conditional Flows Official implementation of the paper DeFlow: Learning Complex Im

Valentin Wolf 86 Nov 16, 2022
A PyTorch port of the Neural 3D Mesh Renderer

Neural 3D Mesh Renderer (CVPR 2018) This repo contains a PyTorch implementation of the paper Neural 3D Mesh Renderer by Hiroharu Kato, Yoshitaka Ushik

Daniilidis Group University of Pennsylvania 1k Jan 09, 2023
Implementation of "Fast and Flexible Temporal Point Processes with Triangular Maps" (Oral @ NeurIPS 2020)

Fast and Flexible Temporal Point Processes with Triangular Maps This repository includes a reference implementation of the algorithms described in "Fa

Oleksandr Shchur 20 Dec 02, 2022
A repository for the paper "Improved Adversarial Systems for 3D Object Generation and Reconstruction".

Improved Adversarial Systems for 3D Object Generation and Reconstruction: This is a repository for the paper "Improved Adversarial Systems for 3D Obje

Edward Smith 188 Dec 25, 2022
Course content and resources for the AIAIART course.

AIAIART course This repo will house the notebooks used for the AIAIART course. Part 1 (first four lessons) ran via Discord in September/October 2021.

Jonathan Whitaker 492 Jan 06, 2023
Unbiased Learning To Rank Algorithms (ULTRA)

This is an Unbiased Learning To Rank Algorithms (ULTRA) toolbox, which provides a codebase for experiments and research on learning to rank with human annotated or noisy labels.

71 Dec 01, 2022
Using machine learning to predict undergrad college admissions.

College-Prediction Project- Overview: Many have tried, many have failed. Few trailblazers are ambitious enought to chase acceptance into the top 15 un

John H Klinges 1 Jan 05, 2022
The code for replicating the experiments from the LFI in SSMs with Unknown Dynamics paper.

Likelihood-Free Inference in State-Space Models with Unknown Dynamics This package contains the codes required to run the experiments in the paper. Th

Alex Aushev 0 Dec 27, 2021
Styled Augmented Translation

SAT Style Augmented Translation Introduction By collecting high-quality data, we were able to train a model that outperforms Google Translate on 6 dif

139 Dec 29, 2022
NEATEST: Evolving Neural Networks Through Augmenting Topologies with Evolution Strategy Training

NEATEST: Evolving Neural Networks Through Augmenting Topologies with Evolution Strategy Training

Göktuğ Karakaşlı 16 Dec 05, 2022