Latent Convolutional Models
Sample resotrations using a Latent Convolutional Model.
Latent Convolutional Models work by parametrizing the latent space of a generator using convolutional neural networks. A schematic can be found below
The Schematic of a Latent Convolutional Model. The smaller ConvNet f (red) is unique to each image is parametrize the latent space of the generator g_theta (magenta) which is common to all images. The input s is fixed to random noise and is not updated during the training process.
Installation Dependencies
- numpy 1.14.3
- pytorch 0.4.0
- tensorboard-pytorch