This script generates new images of cats using the technique of generative adversarial networks (GAN), as described in the paper by Goodfellow et al. The images are enhanced with the laplacian pyramid technique from Denton and Soumith Chintala et. al., implemented as a single G (generator) as described in the blog post by Anders Boesen Lindbo Larsen and Søren Kaae Sønderby. Most of the code is based on facebook’s eyescream project. The script also uses code from other repositories for spatial transformers, weight initialization and LeakyReLUs.
The following images were generated by networks trained with:
The difference between model G32up and G32up-c is simply that G32up-c is about one layer deeper and has more convolution kernels.
256 randomly generated 32×32 cat images. (Model G32up-c)
64 generated 32×32 cat images, rated by D as the best images among 1024 randomly generated ones. (Model G32up-c)
16 generated images (each pair left) and their nearest neighbours from the training set (each pair right). Distance was measured by 2-Norm (torch.dist()). The 16 selected images were the “best” ones among 1024 images according to the rating by D, hence some similarity with the training set is expected. (Model G32up-c)
Training progress of the network while learning to generate color images. Epoch 1 to 750 as a youtube video. (Model G32up-c)
256 randomly generated 32×32 cat images. (Model G32up)
64 generated 32×32 cat images, rated by D as the best images among 1024 randomly generated ones. (Model G32up)
You can read the article in its entirety, on the official website of https://github.com/aleju/cat-generator
Thanks a lot … I hope your thoughts, your geniuses can have as much resonance as possible.