I'm building Autoencoder. In the encoding part I use several strided and diluted convolutions per stage of encoding but I'm wondering if I have to construct the decoding part in a symmetric way.

My thought is the following : if the encoding takes into account all the input scales (with the diluted and strided convolutions) the decoding part should reconstruct all the scales without much effort.

Is the encoding enough to ensure a good reconstruction ?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.