Light

hyyh1314 / bgshadownet Goto Github PK

View Code? Open in Web Editor NEW

20.0 4.0 6.0 5.38 MB

Python 100.00%

bgshadownet's Introduction

Document Image Shadow Removal Guided by Color-Aware Background

Requirements

python 3.6.13
pytorch 1.10.0
numpy==1.19.2
opencv_python_headless==4.5.4.60

TRAIN

You can obtain RDD dataset from here
Modify the config.yaml to set your parameters
You can generate the csv file you need by configuring, running utils/make_dataset.py
Use extractBackground.py to construct the ground-truth background
Training CBENet

python Train_CBENet.py ./configs/model=CBENet/config.yaml

Training BGShadowNet

python Train_BGShadowNet.py ./configs/model=BGShadowNet/config.yaml

Pretrained model

Download the pretrained model trained on RDD dataset.

TEST

python test.py

Cite

  @InProceedings{Zhang_2023_CVPR,
  author  = {Ling Zhang,Yinghao He,Qing Zhang,Zheng Liu,Xiaolong Zhang,Chunxia Xiao},
  title   = {Document Image Shadow Removal Guided by Color-Aware Background.},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  month   = {June},
  year   = {2023}}

bgshadownet's People

Contributors

Stargazers

Watchers

Forkers

nuaa-xsf firewalldragondarkfluid strongenemy piriyabiw liviof nauyihsnehs

bgshadownet's Issues

test in image without gt

how can i use test.py without gt image

About the pretrained model:

The pretrained model you provided is not open, hope this issue could be finished, best wishes.
And this is the situation that i met:

extractBackground.py is so slow ooo

请问作者，extractBackground.py有什么加速方法吗？

Could you please share the RDD dataset and pretrained model

The current Google drive links do not work.

About the test results.

Could you provide the deshadowing results for RDD and Kilger datasets, or could you share your evaluation script? I've noticed that the RMSE, PSNR, and SSIM values I obtained using the model weights you provided are significantly different from the paper. Specifically, on RDD, I obtained RMSE: 6.02, PSNR: 33.33, SSIM: 0.9520, without resizing the input images. Additionally, I found that when using Kilger's original images as input, my 24GB GPU memory is insufficient. Could you please clarify if you perform data splitting or resizing when evaluating this dataset?"

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.