MANGAIRO
MangaIRO

MangaIRO

ID: mangairo // PROJECT
DATE: Apr 2024
PROJECT_OVERVIEW

High-performance GAN-based model using PyTorch for automatic manga colorization, implementing Pix2Pix and CycleGAN architectures.

TECH_STACK
  • > PyTorch [OK]
  • > GAN [OK]
  • > Computer Vision [OK]
  • > Pix2Pix [OK]
SOURCE_CODE
PROJECT_DETAILS

MangaIRO 🎨📖

Automatic Manga colorization using Generative Adversarial Networks (GANs).

Features

  • GAN Architectures: Implements Pix2Pix and CycleGAN for image-to-image translation.
  • Grayscale to Color: Converts detailed manga panels into vibrant color versions.
  • Pytorch Implementation: Optimized training and inference pipeline.

Tech Stack

  • PyTorch: Deep learning framework.
  • GANs: Generative Adversarial Networks.
  • Computer Vision: Image processing and enhancement.