junyanz/CycleGAN

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

12870 stars · 1952 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A Torch (Lua) implementation of CycleGAN and pix2pix for unpaired image-to-image translation using cycle-consistent adversarial networks. It can translate photos to paintings, horses to zebras, and perform style transfer without paired training data.

Use cases

• convert paintings to photos and photos to paintings
• turn horses into zebras in images
• perform unpaired image style transfer
• train image-to-image translation models without paired data
• reproduce CycleGAN and pix2pix research results
• experiment with GAN-based image manipulation

When to choose

• you need unpaired image-to-image translation with the original Torch implementation
• you want to reproduce the ICCV 2017 CycleGAN paper results
• you work in a Torch/Lua deep learning environment

When to avoid

• you want an actively developed PyTorch version - use junyanz/pytorch-CycleGAN-and-pix2pix instead
• you need a maintained production pipeline
• you prefer TensorFlow or Keras implementations

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision deep-learning computer-vision image-processing machine-learning gan cyclegan pix2pix image-to-image-translation style-transfer torch generative-models linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="junyanz/CycleGAN")