Skip to content

Latest commit

 

History

History
54 lines (38 loc) · 2.66 KB

README.md

File metadata and controls

54 lines (38 loc) · 2.66 KB
Raw

ImgRestoration for Divers!

  • This repository is an easy toolkit to restore the underwater image/video using Deep Learning model!
  • It uses the model from the paper Underwater Ranker: Learn Which is better and How to be better [AAAI 2023 Oral]

Get Started

  • Install python>=3.6
  • Git clone this repository by typing this command in the command line:
    • git clone https://github.com/AllenEdgarPoe/UnderwaterImgRestore4Diver.git
  • Create python virtual environment by using the below command line:
    • python -m venv venv
  • Activate the virtual environment:
    • venv\Scripts\activate
  • Install required libraries:
    • pip install -r requirements.txt
  • Download models from models
    • get NU2Net_ckpt.pth and put it in ./checkpoints/NU2Net_ckpt.pth
  • Start the program!
    • python main.py

Functions

  • Convert the single image/video file by clicking Single file tab
  • Convert the whole image/video files by clicking Batch tab.
  • You can set the save directory by designating it through output_path tab. If you don't, the default saving path is ./results
  • Comparison results will be saved in [save_path]/compare
  • Enjoy the beautiful comparison results in the below tab!

Examples

Dialog

image

Reulst (Image comparison)

All of the pictures were taken by me @ Gili.T (Indonesia) KakaoTalk_20230829_131022483_03 KakaoTalk_20230829_131022483_10 KakaoTalk_20230829_131022483 KakaoTalk_20230829_131022483_06 KakaoTalk_20230829_131022483_04 KakaoTalk_20230829_131022483_13

Result (Video comparison)

dive.mp4

Acknowledgement