Skip to content

Romrchp/optml-project

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

71 Commits
 
 
 
 
 
 
 
 

Repository files navigation

CS-439: Image Denoising Using a Chambolle Scheme with Isotropic and Anisotropic Total Variations

The goal of this project is to use the Chambolle dual algorithm with an isotropic and anisotropic total variation regularization to perform denoising on sample images. The results are then be compared according to the choices of the hyperparameters via analysis of some image quality metrics.

Repository Structure

This repository includes :

  • final_notebook.ipynb : Main Python notebook including our new implementation, the tests and the different experiments conducted, as well as the results obtained in the report. Also contains additional important elements not kept in the report due to space management, that follow the flow of the report.

  • background_notebooks : This folder includes several notebooks, that have been used as inspiration for the project as well some additional experiences.

Prerequisites :

The project was coded and ran under the 3.10.8 version, but should run properly on other versions as well.

The main packages to install for this project are :

  • pyproximal : This Python library provides all the needed building blocks for solving non-smooth convex optimization problems using the so-called proximal algorithms.

  • pooch : Pooch is a Python library that can manage data by downloading files from a server (only when needed) and storing them locally in a data cache (a folder on your computer). It is needed in our project to load the 'Ascent' image.

In order to correctly install the latter and the rest of the more 'basic' dependencies, please refer to the requirements.txt file in the root folder of the repository. It contains the necessary dependencies to run our final_notebook.ipynb.

Run instructions

To run final_notebook.ipynb, clone the repository or download the file and run it from your prefered method. This repository does not contain any external .py or ipynb files needed for this notebook to run.

Main references and inspiration for the project :

Code reference :

Contributors :

  • Christopher Cherfan, Quantum Science and Engineering Section,
  • Lenny Del Zio, Quantum Science and Engineering Section,
  • Romain Rochepeau, Section of Life Sciences Engineering.

About

GitHub Repository of the OptML course.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors