I think I may try, yes, to create a 3-view from slanting views.
The only problem I have is that my trigonometric tools take as basis an "isometric" slanting view (like zoomed from great distance). If the source has a strong conical effect (like photograph from short distance, with big wing front and little wing far), I must average and this is not precise at all.
Well, I did that in the past, it is a time-consuming job, not funny. I may try sunday or the sunday after because I have another work to do for another friend (scanning a whole book).