GithubHelp home page GithubHelp logo

ip-derain / nerd-rain Goto Github PK

View Code? Open in Web Editor NEW

This project forked from cschenxiang/nerd-rain

0.0 0.0 0.0 7.37 MB

Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining (CVPR 2024)

Shell 0.02% Python 96.81% MATLAB 3.17%

nerd-rain's Introduction

【CVPR'2024πŸ”₯】Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining

Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining

Xiang Chen, Jinshan Pan, Jiangxin Dong

Nanjing University of Science and Technology

Primary contact: Xiang Chen ([email protected])

πŸ“£ News

  • [24-04-03] The paper is available here.
  • [24-03-20] The source codes, models and results are available.
  • [24-02-27] Our paper has been accepted to CVPR 2024.

πŸ“– Research Website: [link] (A Survey on Image Deraining)

πŸ“Œ Overview

avatar

πŸ”‘ Setup

Type the command:

pip install -r requirements.txt

Install warmup scheduler

cd pytorch-gradual-warmup-lr; python setup.py install; cd ..

🧩 Dataset Preparation

Datasets Download Link
Rain200L Baidu Netdisk (s2yx)
Rain200H Baidu Netdisk (z9br)
DID-Data Baidu Netdisk (5luo)
DDN-Data Baidu Netdisk (ldzo)
SPA-Data Baidu Netdisk (yjow)

πŸ› οΈ Training and Testing

  1. Please download the corresponding datasets and put them in the folder Datasets/.
  2. Follow the instructions below to begin training our model.
bash train.sh

Run the script then you can find the generated experimental logs in the folder checkpoints.

  1. Follow the instructions below to begin testing our model.
python test.py

Run the script then you can find the output visual results in the folder results/.

πŸ€– Pre-trained Models

Models NeRD-Rain-S NeRD-Rain
Rain200L Google Drive / Baidu Netdisk (thga) Google Drive / Baidu Netdisk (ub1n)
Rain200H Google Drive / Baidu Netdisk (twg1) Google Drive / Baidu Netdisk (x45q)
DID-Data Google Drive / Baidu Netdisk (bljv) Google Drive / Baidu Netdisk (n37f)
DDN-Data Google Drive / Baidu Netdisk (10ut) Google Drive / Baidu Netdisk (01m0)
SPA-Data Google Drive / Baidu Netdisk (p7h9) Google Drive / Baidu Netdisk (mhkn)

🚨 Performance Evaluation

See folder "evaluations"

  1. for Rain200L/H and SPA-Data datasets: PSNR and SSIM results are computed by using this Matlab Code.

  2. for DID-Data and DDN-Data datasets: PSNR and SSIM results are computed by using this Matlab Code.

πŸš€ Visual Deraining Results

Methods DualGCN SPDNet Uformer Restormer
Rain200L Baidu Netdisk (v8qy) Baidu Netdisk (y39h) Baidu Netdisk (N/A) Baidu Netdisk (6a2z)
Rain200H Baidu Netdisk (jnc9) Baidu Netdisk (mry2) Baidu Netdisk (N/A) Baidu Netdisk (9m1r)
DID-Data Baidu Netdisk (3gdx) Baidu Netdisk (klci) Baidu Netdisk (4uur) Baidu Netdisk (1hql)
DDN-Data Baidu Netdisk (1mdx) Baidu Netdisk (19bm) Baidu Netdisk (39bj) Baidu Netdisk (crj4)
SPA-Data Baidu Netdisk (lkeb) Baidu Netdisk (dd98) Baidu Netdisk (N/A) Baidu Netdisk (b40z)
Methods IDT DRSformer NeRD-Rain-S NeRD-Rain
Rain200L Baidu Netdisk (v4yd) Baidu Netdisk (hyuv) Baidu Netdisk (tur9) Baidu Netdisk (jdir)
Rain200H Baidu Netdisk (77i4) Baidu Netdisk (px2j) Baidu Netdisk (g05m) Baidu Netdisk (2l0w)
DID-Data Baidu Netdisk (8uxx) Baidu Netdisk (t879) Baidu Netdisk (p2h6) Baidu Netdisk (yerc)
DDN-Data Baidu Netdisk (0ey6) Baidu Netdisk (9vtz) Baidu Netdisk (f10x) Baidu Netdisk (6stc)
SPA-Data Baidu Netdisk (b862) Baidu Netdisk (bl4n) Baidu Netdisk (c6ky) Baidu Netdisk (lhyw)

πŸ”Ž More Results

Methods Models Descriptions
model_MPRNet.py Google Drive Extension to CNN-based UNet, see Table 6.

πŸ‘ Acknowledgement

Thanks for their awesome works (DeepRFT and NeRCo).

πŸ“˜ Citation

Please consider citing our work as follows if it is helpful.

@InProceedings{NeRD-Rain,
    author={Chen, Xiang and Pan, Jinshan and Dong, Jiangxin}, 
    title={Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month={June},
    year={2024}
}

nerd-rain's People

Contributors

cschenxiang avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

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

  • Typescript photo Typescript

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

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. πŸ“ŠπŸ“ˆπŸŽ‰

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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

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

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❀️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.