This Neural Network Paints Depressing Russian Cityscapes

Neurussia is a neural network that paints Russian landscapes based on Soviet architectural heritage. The project was based on the largest Russian social network — VK. I used the famous Nvidia architecture StyleGAN2 with discriminator augmentations. I got around 10k images of Russian landscapes from VK and used a python wrapper for data scrapping. The official paper implementation uses the.tfrecord format to store data for training. When running on Colab this feature can drastically reduce the number of images that you could process.