Skip to content

alessandrodicosola/DeepDeblurring

Repository files navigation

DeepDeblurring

DeepDebluring project for the course [91250] Deep Learning at UniBo

Requirements

  • python 3.x

Conda

  • conda create --name <env> --file requirements_conda.txt
  • use directly requirements_conda.yml within anaconda

Pip

pip install -r requirements_pip.txt

Usage

  • Go to the project folder (where src, report andmodels folders are)
  • activate environment
  • python src/main_cli.py MODEL PARAMS

Documentation

python src/main_cli.py -h
python src/main_cli.py MODEL -h 

Report

The report is report/Report.pdf

The powerpoint presentation is report/powerpoint.pdf

Models

The models implemented are:

The model and weights are save in /models/[MODEL NAME]/[MODEL NAME]_best.h5

About

Survey different models for deep deblurring: UNet, DenseNet, CAESSC, SRNDeblur

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published