Titan AI LogoTitan AI

CycleGAN

12,768
1,975
Lua

Project Description

Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.

CycleGAN: Software that can generate photos from paintings, turn horses into zebras, perform style transfer,

Project Title

CycleGAN — Unpaired Image-to-Image Translation Framework

Overview

CycleGAN is an open-source software that enables image-to-image translation without paired training examples. It's capable of generating photos from paintings, turning horses into zebras, performing style transfer, and more. This project stands out for its ability to learn translations between two domains using only unpaired examples, which is a significant advancement in the field of generative adversarial networks.

Key Features

  • Unpaired image-to-image translation capabilities
  • Cycle-consistent adversarial networks for learning translations
  • Support for various applications like style transfer, object transfiguration, and season change
  • PyTorch implementation available for active development and improved results

Use Cases

  • Art enthusiasts can use CycleGAN to convert paintings into realistic photos.
  • Designers can apply different artistic styles to their images for creative projects.
  • Researchers can experiment with object transformation and season change effects in images.
  • Developers can integrate CycleGAN into applications requiring image manipulation and style transfer.

Advantages

  • Performs image-to-image translation without the need for paired training data
  • Offers a wide range of applications in style transfer and object transformation
  • Actively developed PyTorch version for better performance and results
  • Open-source, allowing for community contributions and improvements

Limitations / Considerations

  • Requires NVIDIA GPU + CUDA CuDNN for optimal performance
  • CPU mode and CUDA without CuDNN may work but are untested
  • The project is written in Lua, which may not be as popular as other languages like Python

Similar / Related Projects

  • pix2pix: A related project that requires paired training examples for image-to-image translation.
  • contrastive-unpaired-translation (CUT): A newer unpaired image-to-image translation model that enables fast and memory-efficient training.
  • BiGAN/ALI: Other methods for image-to-image translation that are included in the CycleGAN package.

Basic Information


📊 Project Information

  • Project Name: CycleGAN
  • GitHub URL: https://github.com/junyanz/CycleGAN
  • Programming Language: Lua
  • ⭐ Stars: 12,738
  • 🍴 Forks: 1,974
  • 📅 Created: 2017-03-30
  • 🔄 Last Updated: 2025-09-12

🏷️ Project Topics

Topics: [, ", c, o, m, p, u, t, e, r, -, g, r, a, p, h, i, c, s, ", ,, , ", c, o, m, p, u, t, e, r, -, v, i, s, i, o, n, ", ,, , ", c, y, c, l, e, g, a, n, ", ,, , ", d, e, e, p, -, l, e, a, r, n, i, n, g, ", ,, , ", g, a, n, ", ,, , ", g, a, n, s, ", ,, , ", g, e, n, e, r, a, t, i, v, e, -, a, d, v, e, r, s, a, r, i, a, l, -, n, e, t, w, o, r, k, ", ,, , ", i, m, a, g, e, -, g, e, n, e, r, a, t, i, o, n, ", ,, , ", i, m, a, g, e, -, m, a, n, i, p, u, l, a, t, i, o, n, ", ,, , ", p, i, x, 2, p, i, x, ", ,, , ", t, o, r, c, h, ", ]



This article is automatically generated by AI based on GitHub project information and README content analysis

Titan AI Explorehttps://www.titanaiexplore.com/projects/cyclegan-86721514en-USTechnology

Project Information

Created on 3/30/2017
Updated on 10/31/2025