A project reproducing the results of CycleGAN paper and making some additional changes to improve its performance.

Illustration by Malak Hassanain

Introduction

I was introduced to the idea “Image to Image Translation” by a Kaggle competition named “I’m Something of a Painter Myself” . The competition was about creating a GAN model that can translate a given set of digital photos to paintings. After going through almost every publicly available notebook in Kaggle and reading many different articles and the CycleGAN paper I was eventually able to get some idea on CycleGANs.

Find my Pytorch implementation here.

Objective

The model should be able to translate any given digital image photos to Monet like paintings. Some other examples will be conversion of a horse…


A simple document scanner using OpenCV, Python and IP Camera

Document Scanning is a process that most of us use everyday. Often I scan the document using my phone and send it to laptop, but this is a long and tedious process which also eats up space on my phone.

So I made a Document Scanner that scans the document using my phone and saves the photo in my laptop directly which is the cropped and warped, finally everything is saved as a single pdf file. The link to my github repository can be found in the end.

Prerequisites :

  • Python Packages : OpenCV, imutils, numpy and Pillow
  • Mobile Application…

Nived

I help computers think.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store