Generative adversarial networks (GAN) are a class of generative machine learning frameworks. A GAN consists of two competing neural networks, often termed the Discriminator network and the Generator network. GANs have been shown to be powerful generative models and are able to successfully generate new data given a large enough training dataset.
Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.
Image-to-image translation with conditional adversarial nets
Modified implementation of DCGAN focused on generative art. Includes pre-trained models for landscapes, nude-portraits, and others.
[CVPR 2016] Unsupervised Feature Learning by Image Inpainting using GANs
Generate cat images with neural networks
A torch implementation of "Pixel-Level Domain Transfer"
Image De-raining Using a Conditional Generative Adversarial Network
Handwritten Chinese Characters Generation
Sharpness-aware Low Dose CT Denoising Using Conditional Generative Adversarial Network
Add colors to black and white images with neural networks (GANs).
Torch implementation of Wasserstein GAN https://arxiv.org/abs/1701.07875
Image completion with Torch
Semi-supervised InfoGAN
Reversing GAN image generation for similarity search and error/artifact fixing
An implementation of context encoders by Deepak Pathak with a remodeled discriminator.
Released June 10, 2014