Image encoding can be performed in multiple ways. In the following sections, we will contrast the performance of vanilla autoencoders, multilayer autoencoders, and convolutional autoencoders. The term auto-encoding refers to encoding in such a way that the original input is recreated with a far fewer number of dimensions in an image.
An autoencoder takes an image as input and encodes the input image into a lower dimension in such a way that we can reconstruct the original image by using only the encoded version of the input image. Essentially, you can think of the encoded version of similar images as having similar encoded values.
