Skip to content

We will visualize the style transfer output produced by monet_generator_model. We take 5 sample images that are photos of beautiful landscapes in the original dataset and feed them to the model.

Notifications You must be signed in to change notification settings

nisrinaishmh/Build-an-Image-Style-Transfer-Tool-using-CycleGANs

Repository files navigation

IBMSkillsNetwork: IBM GPXX0KSEEN - Build an Image Style Transfer Tool to Transform Photos to Monet Paintings with CycleGANs🌈🪄

Artists like Claude Monet are recognized for the unique styles of their works, such as the unique colour scheme and brush strokes. These are hard to be imitated by normal people, and even for professional painters, it will not be easy to produce a painting whose style is Monet-esque. However, thanks to the invention of Generative Adversarial Networks (GANs) and their many variations, Data Scientists and Machine Learning Engineers can build and train deep learning models to bring an artist's peculiar style to your photos. In this project, you will be guided through building a style transfer tool using CycleGAN to "translate" the photos in your album into paintings that are Monet-esque!

Author✍️

Cognitive Class AI - Roxanne Li

Mentee Assignment📝📂

Task for Mentee from IBM Academy Advance AI Mentor @ Infinite Learning Course🎓

Completed Course: Practice to Build an Image Style Transfer Tool to Transform Photos to Monet Paintings with CycleGAN from llm IBM SkillsNetwork🔭🛸

Mentee Info✨

Name: Nisrina Ishmah Mahira🧕

Program: IBM Academy - Advance Al [Kampus Merdeka Batch 6 Tahun 2024]🏫🎉

Tech Stack💻

  • Numpy
  • Pillow
  • Tensorflow
  • Matplotlib
  • Keras