Deep Learning Project — Drawing Anime Face with Simple Segmentation Mask
Demo
This is a short demo. The full video is down below at the bottom of the article.
Objective
The objective of this project is to build a deep learning model to generate Anime face portraits from segmentation masks.
In this project, I will first manually annotate a small set of images. Then I will use data augmentation and a U-Net model to multiply the number of segmentation masks to build a dataset. Finally, I will train a GauGAN model for synthesizing anime faces from segmentation masks.
1. Semantic Segmentation
Semantic segmentation is the process of assigning a label (aka class id) to each pixel in an image. The outcome of it is a segmentation mask which is an array of size height * width with every pixel containing a class id.