Question about Auto-Encoder as Feature extractor

[babbu3682]

I read your paper but you said you used auto-encoder as a feature extractor.

However, as I analyze your code, you use U-net architecture or V-net architecture. which one is right?

I think Auto-encoder is more reasonable because U-net can cheat by using skip connection.

[fhaghighi]

In general, any pre-trained network can be used as the feature extractor. The train_autoencoder.py is parametric; for Auto-encoder, the option --arch Vnet should be used. The network can be trained with other restoration tasks with U-Net as well.