Menu

Better SVM with YIQ

FaceL
2010-01-10
2013-04-24
  • Tom Babbitt

    Tom Babbitt - 2010-01-10

    While heuristically recapitulating the face recognition using PIL and LIBSVM i discovered that the recognition probabilities are much higher if the image is first converted into YIQ color space. With PIL it is;
    matrix = (0.299,0.587,0.114,0,0.596,-0.274,-0.322,0,0.212,-0.523,0.311,0)
    Image = Image.convert("RGB",matrix)

     
  • David S. Bolme

    David S. Bolme - 2010-01-11

    Thanks for the tip.  I will look into it.

     

Log in to post a comment.