Light

wonbeomjang / esrgan-pytorch Goto Github PK

View Code? Open in Web Editor NEW

99.0 2.0 26.0 164.64 MB

re-implementation super resolution pytorch using ESRGAN

License: MIT License

Python 100.00%

esrgan-pytorch's Introduction

ESRGAN-pytorch

This repository implements a deep-running model for super resolution. Super resolution allows you to pass low resolution images to CNN and restore them to high resolution. We refer to the following article.
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

architecture

[Overall Architecture]
[Basic block]

Test Code

python test.py --lr_dir LR_DIR --sr_dir SR_DIR

Prepare dataset

Use Flicker2K and DIV2K

cd datasets
python prepare_datasets.py
cd ..

custom dataset

Make dataset like this; size of hr is 128x128 ans lr is 32x32

datasets/
    hr/
        0001.png
        sdf.png
        0002.png
        0003.png
        0004.png
        ...
    lr/
        0001.png
        sdf.png
        0002.png
        0003.png
        0004.png
        ...

how to train

run main file

python main.py --is_perceptual_oriented True --num_epoch=10
python main.py --is_perceptual_oriented False --epoch=10

Sample

we are in training on this code and train is not complete yet. this is intermediate result.

esrgan-pytorch's People

Contributors

Stargazers

Watchers

esrgan-pytorch's Issues

How to use the src/test.py?

in src, python3 test.py in my PC, no results. and how to use it ?

wanna the weight file

Can you share the trained weight file? My result is not very good

Dimension mismatch error on changing image dimensions

My hr images are 256x256 and lr is 64x64, with batch_size=8 and scale=4

I'm getting this error:
RuntimeError: size mismatch, m1: [8 x 73728], m2: [8192 x 100] at /opt/conda/conda-bld/pytorch_1579022060824/work/aten/src/THC/generic/THCTensorMathBlas.cu:290

损失函数相关

我看到您loss.py中perception_loss与SRGAN中相同，但是ESRGAN中作者指出该损失应该取激活之前的特征进行求得，请问根据您使用的loss情况，超分效果如何？指标是否下降？

Where is the Discriminator?

code exit

when running 'python main.py', this program will be exit right now . and no message output

Output images are completely black

Dear author,
Thank you so much for sharing the simplified code of ESRGAN.

I have used your training code for my own dataset, but testing code produced completely black super resolved images.

Kindly guide.
Best,
Farooq

How much GPU memory is needed for training?

Hi，
I set the batchsize to 1 to train on a RTX 3080, but it prompts that the GPU memory is insufficient to train. So I would like to ask, what GPU are you training on?
Sincerely，
Chison

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.