Originally posted on My Medium. In this article, I will compare and show you the evolution of StyleGAN, StyleGAN2, StyleGAN2-ADA, and StyleGAN3. Note: some details will not be mentioned since I want to make it short and only talk about the architectural changes and their purposes. StyleGAN The purpose of StyleGAN is to synthesize photorealistic/high-fidelity […]
Paper Explained: StyleGAN3 — Alias-Free Generative Adversarial Networks Originally posted on My Medium. The StyleGAN3 paper is pretty hard to understand. In this article, I tried my best to reorganize it and explain it step by step. Hope you understand it better after reading this. If you want to know the difference and evolution of StyleGAN, […]
Paper Explained: SeFa — Closed-Form Factorization of Latent Semantics in GANs Originally posted on My Medium. Motivation The generator in GANs usually takes a randomly sampled latent vector as the input and generates a high-fidelity image. By changing the latent vector , we can change the output image. However, in order to change a specific attribute in the output […]
2 ways to quickly solve the viral puzzle Originally posted on My Medium. Unlike my usual articles, this is not about deep learning or computer vision. This is about a popular web game called Wordle. Since I have been seeing many people sharing their daily Wordle results on social media recently, it makes me curious to […]
Originally posted on My Medium. 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 […]
Originally posted on My Medium. In Style Transfer and some generative networks, special normalization layers such as CIN, and AdaIN are used to control the style of the output image. But how and why does a normalization layer affect image style? It is crucial to understand the rationale behind them in order to know why […]
Originally posted on My Medium. Introduction This is an update on my previous article Anime Illustration Colorization Part 1. If you have not read that, I recommend you to read it first and then come back. After posting the previous article, I have done a few things to try to improve the performance of the model. […]
Originally posted on My Medium. Introduction This project aims to create a deep learning model to learn a mapping from a non-colored anime illustration (sketch) to a colored version. We will use the Pix2Pix GAN to achieve this. 1. Dataset First of all, we need a dataset, a dataset that contains a set of non-colored […]
In this tutorial, I will show you how to convert RGBA images to RGB images in the simplest ways using OpenCV (cv2) and Pillow (PIL). An RGBA image has 4 channels: red, green, blue, and alpha. The alpha channel stores how transparent each pixel is, while an RGB image only has the 3 color channels […]
This tutorial will show you how to read/open an image from a URL using OpenCV (cv2) and Pillow (PIL). Also, we will show you how to read the image from a URL with request headers. If you want to read them into Base64, you may read Image & Base64. 1. Read From a URL OpenCV […]