I'm basing my project on the facerecognizer tutorials, but i'm not sure if i do either. Would recognition still work or would it be severely affected?
I think it works pretty well right now but obviously if i could improve it, great. Or maybe it is a thing that's not needed anymore with the new opencv?
There is a normalize function in them, but i'm not sure if that's the same as equalizing them.
The alignment is done in python, not sure why this wasn't done as part of the program itself. Is there a c++ version of doing this? I'm assuming the alignment has to be done on the pictures before we train them, same thing for equalizehist.
↧