I put together some autoencoder slides to summarise the various autoencoder types, which can be found at the following link (browsed using using the space bar):
These were created using a Jupyter notebook and RISE. You can play with and modify the original notebook here. These slides are heavily based on the keras examples and the keras blog
Extra notes
As an aside, we’ve recently published a paper that applies a U-shaped CNN for the segmentation of the liver from MRI. A link to the paper can be found here.