Geo ref images - dir.fov and cross fov inconsistent?

Hi to everybody.
I ran a photo mission made to be flown with my Iris and gopro at medium resolution using instead a Raspberry Pi cam (see it here: thingiverse.com/thing:753269).
In fact it was the inauguration flight and forgot that I had to update the survey grid to obtain full coverage of the area.

I’m puzzled now by the dir.fov and cross fov parameters in the ‘geo ref images’ menu.

The camera has the up direction pointing ‘forward’ hence I’m guessing that in my case
dir.fof is totally equivalent to vertical fov while cross fos is equivalent to the vertical fov.
Filling them accordingly (the RasPi cam has an H fov of 53° and a V fov of 41.4°) unfortunately the produced Kml image seems a bit less than double size respect to googleearth imagery.
To obtain an image consistent with the google earth location I have to set something around 30° V fov and 23° H fov but I can’t understand why.

Using a mission designed for a different camera lens type shouldn’t be an issue as I couldn’t find any camera related data kept in the log.
Those dir.fov and cross fov should be independent and used just to obtain, together with ‘rotation’ and flight heigth (taken from the log together with gps coordinates) the correct resizing to lay down the images on google earth?
. . . but actually all of this is just personal guessing as I haven’t found any documentatiotion about this.

Is anybody out there able to help me out with this?