Black level normalization and data type

[RafiqueA03]

Hi,

Since Gheler Shi and NUS datasets have been used for this work, I have a couple of questions to ask. It would be nice of you to clear.

1. Since for both datasets black level normalization is required before any processing. For this work, how it has been carried out? For the NUS dataset, black level values and saturation values are given for each camera whereas for Gheler-Shi only a black level value is given.
2. Also as you have mentioned the input type is 16-bit but the images in the Gheler Shi dataset are 12-bit so how images have been converted to 16-bit? Is it done by scaling images to 16-bit using normalization?

Thanks