Eigenfaces Algorithm

Question

I am programming a face recognition program using OpenCV.

When it generates its own functions:

• Do I need to use a large database of unknown persons?
• Do I need to use only photographs of people whom I want my system to recognize?
• Do I need to use both?

I'm talking about generating your own functions, this is a learning step .

And how many photos do I need to use to have decent accuracy? More like 20 or 2000?

thank

Answer 1

" " PCA. , .

( , 10 . ), ( ). - , . , .

20 , 100 , .

Answer 2

, ; - 20 , , , . , .