SNcGAN - Generate Conditional Images
Spectral Norm + Conditional GAN
Adeel Mufti, Biagio Antonelli, Julius Monello
Link to paper: https://arxiv.org/abs/1903.06259
GitHub repo: http://github.com/AdeelMufti/SNcGAN
This is a demonstration of a Generative Adversarial Network with Spectral Normalization that has been conditionally trained on images of oil paintings of faces of people, extracted using OpenCV, from the PainterByNumbers and BAM datasets, with conditioning labels created using Microsoft Face API. Additionally, a SNcGAN trained on the faces of celebrities from CelebA dataset is demonstrated.
The form below generates images from a live TensorFlow model, using the conditional labels chosen.
 Miyato, Takeru, et al. "Spectral normalization for generative adversarial networks." arXiv preprint arXiv:1802.05957 (2018).
This project won 2nd place out of 124 projects in a competition hosted by IBM.
 Mirza, Mehdi, and Simon Osindero. "Conditional generative adversarial nets." arXiv preprint arXiv:1411.1784 (2014).
 Duck, Small Yellow. "Painter by numbers, wikiart.org" Kaggle (2016).
 Wilber, Michael J., et al. "BAM! the behance artistic media dataset for recognition beyond photography." Proc. ICCV. Vol. 1. No. 2. (2017).
 Microsoft Face API
 Liu, Ziwei, et al. "Deep Learning Face Attributes in the Wild" ICCV (2015).
Our conditional GAN architecture