Image-Coloring-using-Autoencoders-and-Deep-Transfer-Learning

This Project is Incomplete and will be work in progress

LinkedIn Post for this Project

$\rightarrow$ Approach

• I tried using Autoencoders and Transfer Learning for this one. I tried VGG16 and InceptionResNetV2 as an encoder/feature extractor layer and a custom decoder layer.

• Although it was coloring my B/W images, but results were not satisfactory. See notebooks

• Some great minds have worked on this problem before, like @EmilWallner and Richard Zhang(@rzhang88), both on Twitter. Their approaches were a bit complex to grasp for me, for now. So, I decided to give this project some rest.

---

$\rightarrow$ Sample Outputs

---

$\rightarrow$ This Project was inspired by Emil Waller

• Coloring-greyscale-images
• Floydhub Blog - Colorizing B&W Photos with Neural Networks
• Emil’s Story as a Self-Taught AI Researcher

$\rightarrow$ For fututre references:

• Kaggle- Image Colorization basic implementation with CNN

• Richard Zhang- Colorful Image Colorization

  • Github

$\rightarrow$ Serrano Academy

• Autoencoders
• Neural Networks & GANs

$\rightarrow$ Code Emporium

• Autoencoders - EXPLAINED
• Variational Autoencoders - EXPLAINED!