how to get the "viewing encoding" image

[HaoLiuHust]

In the paper, it saids "A direct representation of viewing encoding is the view�ing direction, which is computed from the intrinsic param�eters of the polarization camera. If the intrinsic parameters
of the camera are not available, we can also use the 2D co�ordinate (u, v) of each pixel and normalize it to [−1, 1]". But how to "computed from the intrinsic param�eters of the polarization camera."?