Question on the post-processing steps of reconstruction

[ercanburak]

Hi @fedepare, and thanks for this great work.

I have a question about the post-processing steps performed after the image reconstruction network. In the paper, it is stated that the output of the reconstruction network is first passed through an exponential function, and then a min/max normalization is applied. The relevant section from the paper can be seen in below image:

When I checked the codes in this repo, I saw that you applied the min/max normalization (here and here), without using an exponential function.

Is this a mistake in the code repo? I'm guessing that you were also using the exponential function while performing experiments in 4.2. Reconstruction Evaluation section of your paper. Is this correct?

Thanks again for sharing this amazing work!